📉 Add PEFT support for PPOTrainer

[ccs96307]

What does this PR do?

This PR provides an initial implementation of PEFT support for PPOTrainer, targeted towards the policy model. The existing functionality remains fully intact and unaffected by these modifications.

The following tasks have been completed:

• Add peft_config for PPOTrainer
• Add unit test for tests/test_ppo_trainer.py (next: implement unit tests to cover the new peft_config integration)
• Update examples/scripts/ppo/ppo_tldr.py
• Update examples/scripts/ppo/ppo.py

As of now, this implementation does not support quantization or gradient checkpointing.

If there’s anything inappropriate or suboptimal in my implementation, please feel free to let me know—I’m open to all feedback and eager to improve to align with the project standards.

Fixes #1916

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? 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.
• 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.

[qgallouedec]

very nice @ccs96307! looking into details

[HuggingFaceDocBuilderDev]

The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update.

[qgallouedec]

Thanks a lot @ccs96307 for your contribution!