I agree on the problem, not necessarily with the solution.
Let’s go back to the root of the problem.
What types of user problems do we encounter on the forum?
- A device not managed by Gladys that they thought would be managed in Gladys but they can’t manage it.
- A device not discovered by Gladys even though it should be compatible according to the website.
- A user interface bug
If you see other specific cases of posts on the forum, let me know 
In 1., it’s a communication issue: our documentation is not rich enough and the user thought it would work, whereas we do not manage the device. The interface could have warned them by saying « No device detected. Is your device in the list of managed devices? [INSERT LINK TO LIST HERE] ».
In 2., it’s a lack of clarity in the interface. The interface could provide more feedback on what the backend « sees » in « raw ». We need to see service by service how we could make the interface clearer with more detailed status returns of the data actually received by the backend. Example: in the case of Xiaomi, we could display a list of the last 10 messages received by Gladys, in a quasi-raw format.
In case 3., often a simple description of how to reproduce the bug is enough to reproduce it on the developer side.
The problem I see with a solution like « we release a file export of the Gladys instance and the user posts it on the forum » is that it encourages the user to be passive in their debugging, and it puts the user in a « user ↔ customer support » relationship with the forum. The file is likely to be a large block filled with technical jargon that the user will not read (as it is unreadable for a non-initiated person), they will simply copy it to the forum.
The result? The support load on the forum will only increase, and user satisfaction will paradoxically decrease as the user will have to wait for a response from an « expert », an « expert » who will not be available because they will have too many requests to answer.
What I see instead
The user should have, in my opinion, all the keys in hand within the product to see what is wrong, and thus be an actor in the resolution of the problem.
We need to go through each service to identify frequently blocking points, and add more messages (error messages, information) in the interface for each identified blocking point.
For simple understanding issues (what is compatible with Gladys?), we need to better redirect to the documentation instead of the forum so that the user can self-answer and get their answer without waiting 
What do you think?
I know my post may be overly complete for the simple feature you requested, but I think it is crucial to lay the right foundations to avoid hours of unnecessary support where a clearer interface often suffices. It’s simple UX choices like this that will determine whether the v4 can « scale » to thousands of users or not.