Support pass kwargs to cogvideox custom attention processor

[huanngzh]

What does this PR do?

Pass kwargs to custom attention processor is preferable when implementing layout control, attention replacing or introducing side effect to attention computation. Now this feature is supported by sd, sdxl, FLUX, sd3 series, but not CogVideoX. This PR fix the problem, and may support future training-free experiments in CogVideoX series models.

Example code:

classCustomCogVideoXAttnProcessor2_0:
def__init__(self, **custom_args):
super().__init__()
# some initializationdef__call__(self, ..., added_argu=None) ->torch.FloatTensor:
# custom attention forward passcustom_processor=CustomCogVideoXAttnProcessor2_0(**custom_args)
pipe.transformer.set_attn_processor(custom_processor) # cogvideox pipelineattention_kwargs= {"added_argu": 100} # prepare passing added_argu to attention processorpipe.transformer(..., attention_kwargs=attention_kwargs) # custom transformer forward pass

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.

Hello @yiyixuxu@sayakpaul, I just want to do the same to CogVideoX as to FLUX in #9517 and to SD3 in #9818

[a-r-r-o-w]

Thanks!

[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]

thanks!