The database migration helpfully duplicated all of the user records, despite having a constraint that prevents such a thing. fortunately, SQL is a hell of a drug and I could fix that fast once I realized what was happening.
I believe that was causing most of the 500 errors we were seeing - I am not prepared to say that all of them are fixed, but it is better.
On the image front, I accidentally/unknowingly stopped the copy of the images from the old server to the shiny new wasabi storage bucket. With the way Wasabi transfers work, the 20GB of media should finish sometime in March of 2026 (give or take). Seriously, though, it'll take a while and then should be good to go.
Let me know if you see more issues.
Just remember that there are still a thundering herd of bugs in kbin still that the community is working through and the issues I fixed here were those due to load.