What does this PR do?

This PR adds a conversion script convert_diffusers_to_cogvideox.py that enables converting models from the Diffusers format back to the original CogVideoX format. This complements the existing convert_cogvideox_to_diffusers.py script, allowing for bidirectional conversion between the two formats.

The script provides functionality to:

• Convert Transformer models from Diffusers to CogVideoX format
• Convert VAE models from Diffusers to CogVideoX format
• Support both Image-to-Video and regular CogVideoX models
• Handle different model versions (1.0 and 1.5)
• Support FP16, BF16, and FP32 precision

Fixes #10076

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.
  • Approved in Do we have any script covert from hf format to orginal format? #10076
• Did you make sure to update the documentation with your changes? Here are the
  documentation guidelines, and
  here are tips on formatting docstrings.
  • Other conversion scripts do not have any documentation
• Did you write any new necessary tests?
  • Other conversion scripts do not have any tests, can commit tests if needed