To ensure that an incident like what happened on October 24 with the original forum doesn't happen again, I decided to write something in Python that scrapes the forums as a proof of concept.
Here's what an archived thread's posts file usually look like. Fun fact, this particular thread's posts file is about 1 MB in total size. No other thread has managed to reach this size yet.
What worried me at first was the endless scrolling system, which would make archiving threads slightly more difficult. However, it turns out that Flarum's API is an archivist's wet dream and could be used to save threads in a much smaller and faster way than the traditional method of saving HTML files, completely bypassing the endless scrolling system in the process. Nice one, Flarum!
If we can't move the database from Flarum to WoltLab very cleanly, I think this could come in handy for a public "archive" of some sort. However, this apparently requires the permission of every single person so I'm not sure it would be practical.