Check if q.shape[0] * q.shape[1] is 2**18 or larger and use the lower memory usage MPS optimization if it is. This should prevent most crashes that were occurring at certain resolutions (e.g. 1024x1024, 2048x512, 512x2048).
Also included is a change to check slice_size and prevent it from being divisible by 4096 which also results in a crash. Otherwise a crash can occur at 1024x512 or 512x1024 resolution.
When saving training results with torch.save(), an exception is thrown:
"RuntimeError: Can't call numpy() on Tensor that requires grad. Use tensor.detach().numpy() instead."
So for MPS, check if Tensor.requires_grad and detach() if necessary.
This prevents scenarios where text without line breaks will start overlapping with each other when generating X/Y plots. This is most evident when generating X/Y plots with checkpoints, as most don't contain spaces and sometimes include extra information such as the epoch, making it extra long.
This fixes the following error when SD attempts to start with a deleted checkpoint:
```
Traceback (most recent call last):
File "D:\Web\stable-diffusion-webui\launch.py", line 295, in <module>
start()
File "D:\Web\stable-diffusion-webui\launch.py", line 290, in start
webui.webui()
File "D:\Web\stable-diffusion-webui\webui.py", line 132, in webui
initialize()
File "D:\Web\stable-diffusion-webui\webui.py", line 62, in initialize
modules.sd_models.load_model()
File "D:\Web\stable-diffusion-webui\modules\sd_models.py", line 283, in load_model
checkpoint_info = checkpoint_info or select_checkpoint()
File "D:\Web\stable-diffusion-webui\modules\sd_models.py", line 117, in select_checkpoint
checkpoint_info = checkpoints_list.get(model_checkpoint, None)
TypeError: unhashable type: 'list'
```
Recently, the option to do latent upscale was added to txt2img highres
fix. This feature runs by scaling the latent sample of the image, and
then running a second pass of img2img.
But, in this edition of highres fix, the image and parameters cannot be
changed between the first pass and second pass. We might want to do a
fixup in img2img before doing the second pass, or might want to run the
second pass at a different resolution.
This change adds the option for img2img to perform its upscale in latent
space, rather than image space, giving very similar results to highres
fix with latent upscale. The result is not exactly the same because
there is an additional latent -> decoder -> image -> encoder -> latent
that won't happen in highres fix, but this conversion has relatively
small losses
