Breaks with tensors on XPU device

[a-One-Fan]

Partially a Comfy issue, but I figured I might as well report this here instead.

import torch
import comfy_kitchen as ck
x = torch.randn(128, 256, device='xpu', dtype=torch.bfloat16)
ck.quantize_per_tensor_fp8(x, torch.Tensor((1.0,)).to('xpu'))

Errors out as XPU devices are not allowed for eager.

Stack trace

python -c "import torch; import comfy_kitchen as ck; x = torch.randn(128, 256, device='xpu', dtype=torch.bfloat16); ck.quantize_per_tensor_fp8(x, torch.Tensor((1.0,)).to('xpu'))"
Traceback (most recent call last):
  File "<string>", line 1, in <module>
  File "C:\Users\Vikto\Desktop\Tools\Github_Repos\comfytest2\Comfy_Intel\cenv\lib\site-packages\comfy_kitchen\__init__.py", line 65, in quantize_per_tensor_fp8
    return torch.ops.comfy_kitchen.quantize_fp8(x, scale, dtype_code)
  File "C:\Users\Vikto\Desktop\Tools\Github_Repos\comfytest2\Comfy_Intel\cenv\lib\site-packages\torch\_ops.py", line 1243, in __call__
    return self._op(*args, **kwargs)
  File "C:\Users\Vikto\Desktop\Tools\Github_Repos\comfytest2\Comfy_Intel\cenv\lib\site-packages\torch\_library\autograd.py", line 111, in autograd_impl
    result = forward_no_grad(*args, Metadata(keyset, keyword_only_args))
  File "C:\Users\Vikto\Desktop\Tools\Github_Repos\comfytest2\Comfy_Intel\cenv\lib\site-packages\torch\_library\autograd.py", line 40, in forward_no_grad
    result = op.redispatch(keyset & _C._after_autograd_keyset, *args, **kwargs)
  File "C:\Users\Vikto\Desktop\Tools\Github_Repos\comfytest2\Comfy_Intel\cenv\lib\site-packages\torch\_ops.py", line 836, in redispatch
    return self._handle.redispatch_boxed(keyset, *args, **kwargs)  # type: ignore[return-value]
  File "C:\Users\Vikto\Desktop\Tools\Github_Repos\comfytest2\Comfy_Intel\cenv\lib\site-packages\torch\_library\custom_ops.py", line 344, in backend_impl
    result = self._backend_fns[device_type](*args, **kwargs)
  File "C:\Users\Vikto\Desktop\Tools\Github_Repos\comfytest2\Comfy_Intel\cenv\lib\site-packages\torch\_compile.py", line 53, in inner
    return disable_fn(*args, **kwargs)
  File "C:\Users\Vikto\Desktop\Tools\Github_Repos\comfytest2\Comfy_Intel\cenv\lib\site-packages\torch\_dynamo\eval_frame.py", line 929, in _fn
    return fn(*args, **kwargs)
  File "C:\Users\Vikto\Desktop\Tools\Github_Repos\comfytest2\Comfy_Intel\cenv\lib\site-packages\torch\_library\custom_ops.py", line 377, in wrapped_fn
    return fn(*args, **kwargs)
  File "C:\Users\Vikto\Desktop\Tools\Github_Repos\comfytest2\Comfy_Intel\cenv\lib\site-packages\comfy_kitchen\backends\eager\quantization.py", line 211, in _op_quantize_fp8
    impl = registry.get_implementation("quantize_per_tensor_fp8", kwargs=kwargs)
  File "C:\Users\Vikto\Desktop\Tools\Github_Repos\comfytest2\Comfy_Intel\cenv\lib\site-packages\comfy_kitchen\registry.py", line 269, in get_implementation
    selected_backend = self.get_capable_backend(func_name, kwargs)
  File "C:\Users\Vikto\Desktop\Tools\Github_Repos\comfytest2\Comfy_Intel\cenv\lib\site-packages\comfy_kitchen\registry.py", line 233, in get_capable_backend
    raise NoCapableBackendError(func_name, failures)
comfy_kitchen.exceptions.NoCapableBackendError: No backend can handle 'quantize_per_tensor_fp8': eager: x: device xpu not in {'cuda', 'cpu'}

comfy-kitchen/eager/__init__.py, line 30

all_devices = frozenset({"cuda", "cpu"}) -> all_devices = frozenset({"cuda", "cpu", "xpu", "mps", "others idk"}) ? And maybe triton too?