tl;dr: You do seem to have a problem at the table, but whether or not it's "metagaming" is largely beside the point. Your DM may not be flexible or experienced enough to accommodate your correct guesses, and you may be using the criminal background too expansively.
It's probably not metagaming...
Using a PC's backgrounds and specializations to address in-game obstacles is what those backgrounds and specializations are for. Even if it's a stretch to say that a character may be able to use a particular feat, feature, tool, or ability to deal with a challenge successfully, that doesn't mean that a character wouldn't try to use what they have to get what they need.
Wizards tend to look for magical solutions to problems and Rogues tend to look for sneaky solutions. That's what the game is. A former criminal spent a fair amount of time thinking that crime and criminals are the right way to approach issues, and developing the skills, sensibilities, and contacts to do so. It is 100% appropriate for such a character to try to use the traits they think might apply in situations where they think those traits would help.
...but it might be.
Your question lists a lot of things that you think a criminal "might know". That's fair! But people may not be on the same page with the details.
- If your character's criminal past was mostly as a pickpocket, as
opposed to a smuggler, it's tenuous to say that they would know all
about secret compartments used to smuggle stolen goods.
- If your character was an enforcer for a criminal gang, they might
know a lot about how such gangs operate and could find a reliable
underground informant, but might know nothing at all about counterfeiting or money
laundering.
- If your character was an experienced smuggler they might know about
moving goods discreetly, but that might not be much help in finding a
specific smuggling route through a swamp they otherwise don't even
know exists.
- A hired killer might be great at plotting and carrying out
surreptitious assassinations, but might be totally clueless about how
to carry out a con or a profitable fraud.
Trying to say that your character has an encyclopedic, or even generally practical, knowledge about any illegal activity of any type places a lot of plot-mechanical emphasis on a background that costs nothing to take at character creation. This is made slightly more intense by your statement that you are playing this character "smart and clever", when the Ability you deem the relevant one (Intelligence) is only a 12 (above average intelligence, but not obviously a character smart enough to out-think a large, sophisticated, and coordinated secret society).
What makes it seem potentially metagamey is that you are using a single, non-mechanically relevant phrase on the character sheet to announce automatic successes on the recurring element of HotDQ's plot. If your guesses weren't generally right then that wouldn't be much of a problem, but since they are the lack of mechanical interaction between the background and gameplay leaves few tools for the DM to affect your ability to skip the investigation and mystery elements of the module's plot. This is especially true if your DM is running the game strictly by the book.
Whether or not this is actually metagaming, the true problem is almost certainly that your correct guesses are cutting out gameplay segments your DM intended and the other players expect. And that content isn't being replaced by anything, nor is your DM capable of (or perhaps interested in) limiting your ability to do so in-game. Hence the blanket demand that you "stop metagaming".
Your character's criminal experience may be poorly related to activities of the dragon cult.
Some of the problem may be with the module itself. The dragon cultists are almost certainly not anything like any criminal groups your character's background may represent. This is not an organized crime gang, where you have the (movie-style) Italian mafia, the Russian mob, and then oh yeah those dragon cultists. They are a secret society with specific, near-term goals and enormous resources-- they do not
burgle a bunch of houses in Greenest, they besiege and rob the entire city!
It's not clear to me that any amount of history as a pickpocket would give you any insight into the workings, general or specific, of an organization like the cult. They're not trying to move a couple of kilograms of drugs past one customs checkpoint. They are stealing towns' worth of valuable goods and moving an army around the region.
It's also my feeling that Hoard of the Dragon Queen is heavily focused on both the mystery of what the cult is doing (and why), along with their being generally one or two steps ahead of the players. But the cult is also central to the plot, and there aren't a whole lot of curve balls or cunning schemes around what it does-- accurate guesses aren't all that hard to make.
Getting too much of a leg up on them can take the whole campaign off-book and completely obviate the sequel Rise of Tiamat. It's on your DM to handle that, and it certainly can be done, but not every DM is equally ready and capable of doing it well.
What to do?
There are a few things I've seen (and done) which might help ease these problems. The first and most important is simply to talk to your DM about the issue so that you can be on the same page about why they think you're metagaming, and why you think you're not. This is ideal because it helps ensure that you'll be talking about the problem you actually have, and not solving one you don't but which happens to have similar features.
- Define your character's criminal past more precisely
As above, "criminal" is not a well-defined area of knowledge. This can help define how your character actually uses their background to address current problems:
A former smuggler might be able to piece together information about how so much stolen wealth is being moved around without notice, or be able to identify what things might need to be in place to even try to do so. A petty burglar might know how to find a local fence, who might have some information on an unusual volume of illicit goods moving through a city. A former enforcer or assassin might be able to find the sorts of places where the cult's mercenaries are likely to hang out and get information from them. It is less plausible that a character would be equally capable of all three of these simply by virtue of having been a "criminal".
Knowledge might also have some geographic limitations. A person expert in navigating the criminal underworld in Neverwinter doesn't necessarily have contacts in or information about the underworld in Baldur's Gate.
Importantly, details like this round out your character rather than being a poorly defined Swiss army knife which you, the player, use to back up guesses you probably could have made without your character having that background.
- Ask your DM more frequently if your criminal background applies rather than announcing that it definitely does
It can be irritating to a DM for players to declare that, since they have a "hammer" trait, every problem is some variety of nail. It's not for a player to say that an NPC organization functions in some particular way, let alone that their PC has any deep understanding of or insight into that organization. It's similar to asking your DM if you can make a specific roll. That happens, and often isn't a big deal, but it's not the player's job to decide what is possible, what skills apply, or when opportunities to roll come up.
Asking your DM if your character's background gives them any particular insight shows that you understand the DM's role in the game and leaves the game world clearly in their hands. It also gives them the opportunity to organize the relevant information and place it in context, such as identifying when a roll might be possible (and what it might be) for your character to identify and process relevant information.
- Present ideas more as a suggestion that you look for supporting evidence, rather
than declaring what NPCs are probably doing without any such evidence
This one is a pretty big deal. For a player, correctly guessing the cult's activities isn't a huge strain. But would your characters go all-in on such a guess in-setting? What things might they do to figure out if a guess is worth following up on so that they don't waste their time on an idea that goes nowhere?
Asking a gate guard if they've noticed anything odd about wagons moving through the city might gain you a clue that some looked like they were carrying heavier loads than their shipping manifests suggested. That's a great sign that there might be something odd going on, and is much better in-game than suggesting that the cult is definitely using secret compartments in wagons to smuggle items because they've stolen a lot, and you've seen criminals do it in the past.
It also gives the DM an easier opportunity to run the story in an interesting way, rather than having challenges be trivially overcome at your declaration.
- Consider having the criminal background enable rolls rather than automatically solving problems
This is a follow-on to the above suggestion. It's very sensible to use a background to provide more opportunity for rolls than would exist without it. This keeps the knowledge "on the table", by which I mean constrained by your character's actual abilities.
Rather than saying "my character was a criminal, so they know what scheme would be best in this situation and exactly how to carry it out", you might ask your DM "have I seen anything in my criminal past that resembles this situation?" and then the DM can call for a History check. PCs without the criminal background wouldn't even have a chance to make such a check because they have no criminal experience.
The DM can then provide you with the information that relates to your current obstacle.
- Give up on the idea that there is a single "criminal thing to do",
and the associated idea that knowing such a thing makes your
character an expert on it
Knowing how a pin-and-tumbler lock works doesn't actually give you the ability to definitely pick one; you still need practice to learn the skills involved in picking locks. Knowing that a door is sealed shut by a magic spell, and that locks do something similar, doesn't make you capable of using thieve's tools to defeat the magic spell keeping the door shut just because it's "lock-like" and your character knows how to pick locks.
Defining what your character's criminal background actually covers will help a lot with this, but the core idea is that it isn't an expertise in all things criminal. Your character wasn't necessarily a skilled or successful criminal.
From what you describe, this is the aspect that would irk me the most as a DM. Backgrounds are tools that allow me to make the game more engaging and interesting for players, not tools that make in-game challenges irrelevant. It's true that a "good" DM can arrange things to get the former and avoid the latter, but you can still help out with it.
The more problems you want your background to help you solve, the more detailed that background should be. That way the DM can weave it into the story rather than have it skip or destroy parts of a story.