Fix mixed torch.Tensor and DTensor in generate when use FSDP2 + LoRA

[Xiao-Chenguang]

What does this PR do?

model.generate raise an Fix mixed torch.Tensor and DTensor when FSDP2 and LoRA is applied.

The trainer class is the base class for trl trainers like PPOTrainer and GRPOTrainer in which model.generate is called during the training loop.
When FSDP2 is used, all gather and reshard is automatically done for forward method but not for generate.
This lead to the error above.

To fix it, there is a method register_fsdp_forward_method for torch to manage the DTensor for us.
By registering generate, we can get ride of the error.

generate is not properly supported under FSDP2 accelerating. This PR try to fix this error.

Fixes #42417 (issue)

Before submitting

• This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case).
• Did you read the contributor guideline,
  Pull Request section?
• Was this discussed/approved via a Github issue or the forum? Please add a link
  to it if that's the case.
• Did you make sure to update the documentation with your changes? Here are the
  documentation guidelines, and
  here are tips on formatting docstrings.
• Did you write any new necessary tests?

Who can review?

Anyone in the community is free to review the PR once the tests have passed. Feel free to tag
members/contributors who may be interested in your PR.

[Rocketknight1]

cc @3outeille maybe?

[3outeille]

taking a look today

[3outeille]

lgtm but modify to this

if self.is_fsdp_enabled:
	self.model = self.model_wrapped = model
    # Fix `got mixed torch.Tensor and DTensor` error in model.generate() for FSDP2 with LoRA
    dist.fsdp.register_fsdp_forward_method(self.model, "generate")

[Xiao-Chenguang]

lgtm but modify to this

if self.is_fsdp_enabled:
	self.model = self.model_wrapped = model
    # Fix `got mixed torch.Tensor and DTensor` error in model.generate() for FSDP2 with LoRA
    dist.fsdp.register_fsdp_forward_method(self.model, "generate")

Updated