This PR includes:

Add FiT model (training pipeline / inference pipeline / SwiGLU / 2D RoPE / VisionNTK).
The model is under training and the result seems promising.
Move ImageNet Dataset in DiT to separate file for better look.
Add --imagegrid support in DiT to show images in grid format.
Align the CenterCrop in DiT with the official repo.
Move some universal functions in models/dit.py to utils.py for easier reuse for other models.
Refactor and add positional embedding (2D sinusoidal embedding and 2D Rotatery Embedding) in models/modules, for easier reuse for others models.
Minor formating in DiT folder

TODO:

What does this PR do?

Fixes # (issue)

Adds # (feature)

Before submitting

[ ] This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case).
[x] Did you read the contributor guideline?
[x] Did you make sure to update the documentation with your changes? E.g. record bug fixes or new features in What's New. Here are the documentation guidelines
[x] Did you build and run the code without any errors?
[ ] Did you report the running environment (NPU type/MS version) and performance in the doc? (better record it for data loading, model inference, or training tasks)
[ ] Did you write any new necessary tests?

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.

@xxx