[cogview4][feat] Support attention mechanism with variable-length support and batch packing

[OleehyO]

What does this PR do?

This PR is primarily aimed at adding native multi-resolution + packing training support to CogView4 to better meet user needs. We have conducted relevant tests in this PR.

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?
• Did you read our philosophy doc (important for complex PRs)?
• 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.

@a-r-r-o-w
@zRzRzRzRzRzRzR

[yiyixuxu]

let me know what you think too @a-r-r-o-w
I wonder if it would be easier to make a new attention processor for this?

[OleehyO]

@a-r-r-o-w It has been renamed to CogView4TrainingAttnProcessor, and a bug in the original CogView4AttnProcessor has been fixed. We use the same naming format in both CogView4AttnProcessor and CogView4TrainingAttnProcessor. Please take a look

[OleehyO]

I have changed back to using attention_kwargs in CogView4Transformer2DModel to pass attention parameters.

[a-r-r-o-w]

Thanks for addressing the comments!

[yiyixuxu]

@bot /style

[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.

[yiyixuxu]

@bot /style

[github-actions]

Style fixes have been applied. View the workflow run here.