Break the dependency between torch.nn and torch.distributed

[fegin]

🚀 The feature, motivation and pitch

We are seeing several import issues when compiling distributed modules. And the root cause is there are circular dependencies between torch.nn and torch.distributed. Some examples:

1. torch.nn will import DDP, which will rely on many torch.distributed modules.
2. torch.nn will import pytorch/torch/_jit_internal.py, which rely on torch.distributed.rpc.

While right now we work around these circular dependency issues, but these issues recurrently happen when compiling the distributed modules. We need to ensure lazy dependencies on torch.distributed for torch.nn

Alternatives

No response

Additional context

No response

cc @mrshenli @pritamdamania87 @zhaojuanmao @satgera @gqchen @aazzolini @osalpekar @jiayisuse @H-Huang @kwen2501 @awgu @penguinwu @XilunWu @wanchaol @fduwjj @wz337 @tianyu-l @wconstab @yf225 @chauhang @d4l3k