Add TPU/XLA devices support for ComfyUI

[radna0]

This PR adds support for TPU/XLA devices in ComfyUI. users can enable by adding the command line arg --xla

• Update README about support for TPU/XLA devices.
• Detects TPU/XLA device, and memory information.
• Add torch_xla.sync()/xm.mark_step() for the callback after every sampling step when using TPU devices. This is needed to execute the XLA current graph. *Note: Torch XLA sync is non blocking by default.

Tested on TPU v3 with ComfyUI-MochiWrapper models.

a reference image with sampling settings/seed/etc.