fix: enable unet_3d_condition to support time_cond_proj_dim

[yhZhai]

What does this PR do?

This PR adds an argument time_cond_proj_dim to UNet3DConditionModel (the same as UNet2DConditionModel), and thus it enables the time_embedding module to create a cond_proj module.

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.

[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.

[sayakpaul]

Thanks for the PR. Is this for a use case? Just curious.

[yhZhai]

Thanks for the PR. Is this for a use case? Just curious.

Hi, I encountered this problem when trying to replace the UNet2DConditionModel with UNet3DConditionModel in consistency model training example. In the original code, time_cond_proj_dim is passed to the 2D UNet, and is used to create a new linear layer here. In this way, the 2D consistency model can takes the guidance scale as a conditional input.

This PR enables the consistency model training example to support 3D Unet.

[sayakpaul]

I am okay with the change. Thanks!

[yiyixuxu]

thanks!

[sayakpaul]

@yhZhai please ping us once the CI run is complete.

[yhZhai]

@sayakpaul @yiyixuxu @DN6 Hi all, the CI run has completed. Could you please take a look? Thank you!

[sayakpaul]

Sorry this needs to be run again because previously, the CI was broken (for an internal mistake). Should hopefully be fixed 🤞

[yhZhai]

@sayakpaul Hi, could you please take a look at the CI test result? Thanks!

[sayakpaul]

Thank you for your contributions!