OK - it's loading so that's not unexpected. However my previous experience was that the same thing would happen to the UI at all sorts of surprising moments when other processes were ongoing, such as linting or using the debugger. Kind of hard to reproduce without a large project at hand though.

Then I had to stop testing and SIGKILL it because the whole application got into an infinite loop of NPEs in the UI event loop of all places!

I too experienced using intelliJ for many years and this seems largely in line with what I remember.

Part of them because i didn't fully understand Swing when i began the project and partly because synchronizing GUI's to another live process is a quite error prone task.

A debugger process is almost certainly not living in the same thread as the Swing UI of IDEA so this almost surely more of a thread synchroniztion issue rather than Swing issue. (You really don't watch to do any cross-thread touching with Swing apart from for a very limited set of components built for this).

In intelliJ it was like whenever the UI was not immediately responding to mouse events, it would often just lose the event instead of it being correctly processed once the UI thread was freed up.

In the case of IntelliJ there is probably some of their code accepting the event and then probably has some ad-hoc message passing by setting a field that is polled by the receiver thread under the assumption that the receiver always runs faster than a human can act, however as soon as that assumption is broken by some pause you get what you describe.