SeaORM Migration

[aaronleopold]

The database client Stump uses is officially no longer maintained. While the client is otherwise in an OK state, i.e. it isn’t broken and still gets the job done, I believe it be in the project’s best long-term interest to move off of it as to not lag or get stuck if ever we hit a wall with the library.

What this means for Stump is investigating alternatives and investing the time to port all of the existing database interactions and logic to whatever is deemed the best choice. More plainly, I’m going to freeze any non-critical, backend-specific issues and/or features so I can focus on this migration. I’m on the optimistic side of things, and actually feel this could be net positive. While Prisma had a really good developer experience, there were a few inefficiencies that will be good to remove (e.g., #482). This effort has already been started with good progress, SeaORM is the current successor to Prisma and I've knocked out about 70-80% of the core crate's Prisma-to-SeaORM effort. I'm hopeful it won't take more than a couple of months to fully migrate.

This won’t affect frontend-specific features or general critical bugs, I’ll be sure to continue working on those when they get reported. A release cycle will continue, albeit probably at a lower frequency. I’ll aim to promote the current nightly set of changes in the coming weeks, and the current experimental to nightly shortly after.

I will greatly appreciate anyone willing to help out with this migration, as it will be quite involved and time consuming. Please reach out on Discord (or this discussion) and we can coordinate efforts there