Support gradient checkpointing in Qwen2VL ViT

[li-plus]

What does this PR do?

Support gradient checkpointing for Qwen2VL ViT part. The current implementation in main branch only supports gradient checkpointing in language part. This PR further supports checkpointing vision encoder.

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.

cc @ArthurZucker, @amyeroberts, @qubvel