more careful loading of model weights (eliminates some issues with checkpoints that have weird cond_stage_model layer names)

switching time reduced from ~1500ms to ~280ms

load models to VRM instead of RAM (for machines which have bigger VRM than RAM such as free Google Colab server)

support EMA weights in processing (????)

Yep.

Fix gfpgan_model_arch requirement(s).
Add Upscaler base class, move from images.
Add a lot of methods to Upscaler.
Re-work all the child upscalers to be proper classes.
Add BSRGAN scaler.
Add ldsr_model_arch class, removing the dependency for another repo that just uses regular latent-diffusion stuff.
Add one universal method that will always find and load new upscaler models without having to add new "setup_model" calls. Still need to add command line params, but that could probably be automated.
Add a "self.scale" property to all Upscalers so the scalers themselves can do "things" in response to the requested upscaling size.
Ensure LDSR doesn't get stuck in a longer loop of "upscale/downscale/upscale" as we try to reach the target upscale size.
Add typehints for IDE sanity.
PEP-8 improvements.
Moar.

support saving as half

-Update the name of the model feilds
-Update the associated variable names

Hook the model loader into the SD_models file.
Add default url/download if checkpoint is not found.
Add matching stablediffusion-models-path argument.
Add message that --ckpt-dir will be removed in the future, but have it pipe to stablediffusion-models-path for now.
Update help strings for models-path args so they're more or less uniform.
Move sd_model "setup" call to webUI with the others.
Ensure "cleanup_models" method moves existing models to the new locations, including SD, and that we aren't deleting folders that still have stuff in them.