TorchEmbedding cannot be deep copied #1131

Traceback (most recent call last):
  File "/Users/jackmin/Jina/docarray/oink.py", line 16, in <module>
    a.copy(deep=True)
  File "pydantic/main.py", line 657, in pydantic.main.BaseModel.copy
  File "pydantic/main.py", line 612, in pydantic.main.BaseModel._copy_and_set_values
  File "/Users/jackmin/miniconda3/envs/docarray/lib/python3.10/copy.py", line 146, in deepcopy
    y = copier(x, memo)
  File "/Users/jackmin/miniconda3/envs/docarray/lib/python3.10/copy.py", line 231, in _deepcopy_dict
    y[deepcopy(key, memo)] = deepcopy(value, memo)
  File "/Users/jackmin/miniconda3/envs/docarray/lib/python3.10/copy.py", line 153, in deepcopy
    y = copier(memo)
  File "/Users/jackmin/miniconda3/envs/docarray/lib/python3.10/site-packages/torch/_tensor.py", line 100, in __deepcopy__
    return handle_torch_function(Tensor.__deepcopy__, (self,), self, memo)
  File "/Users/jackmin/miniconda3/envs/docarray/lib/python3.10/site-packages/torch/overrides.py", line 1534, in handle_torch_function
    result = torch_func_method(public_api, types, args, kwargs)
  File "/Users/jackmin/Jina/docarray/docarray/typing/tensor/torch_tensor.py", line 225, in __torch_function__
    return super().__torch_function__(func, types_, args, kwargs)
  File "/Users/jackmin/miniconda3/envs/docarray/lib/python3.10/site-packages/torch/_tensor.py", line 1279, in __torch_function__
    ret = func(*args, **kwargs)
  File "/Users/jackmin/miniconda3/envs/docarray/lib/python3.10/site-packages/torch/_tensor.py", line 185, in __deepcopy__
    raise RuntimeError(
RuntimeError: The default implementation of __deepcopy__() for non-wrapper subclasses only works for subclass types that implement new_empty() and for which that function returns another instance of the same subclass. You should either properly implement new_empty() for your subclass or override __deepcopy__() if it is intended behavior for new_empty() to return an instance of a different type.