Add Mac OS support for Apple Silicon (#770)

* Made an install script and auto updates env for mac

* GPU sensors and initial training working for MAC. Still WIP.

* Switch dataloader to single threaded until I can work around some mac pickeling issues.

* Get quantization working on mac

* Fix mac exclusive imports so they don't break other builds.

* Add mac instructions to the UI

toolkit/models/base_model.py

@@ -41,6 +41,7 @@ from torchvision.transforms import functional as TF
 from toolkit.accelerator import get_accelerator, unwrap_model
 from typing import TYPE_CHECKING
 from toolkit.print import print_acc
+from toolkit.basic import flush
 
 if TYPE_CHECKING:
     from toolkit.lora_special import LoRASpecialNetwork
@@ -90,11 +91,6 @@ class BlankNetwork:
         pass
 
 
-def flush():
-    torch.cuda.empty_cache()
-    gc.collect()
-
-
 UNET_IN_CHANNELS = 4  # Stable Diffusion の in_channels は 4 で固定。XLも同じ。
 # VAE_SCALE_FACTOR = 8  # 2 ** (len(vae.config.block_out_channels) - 1) = 8