FlashVideo <> Diffusers

### Model/Pipeline/Scheduler description

There is a brilliant new two-stage video model released under the MIT license: FlashVideo.

Would be amazing to see this in Diffusers!

https://github.com/user-attachments/assets/b53e2a39-3127-4027-a1b2-9f435d20da60

The left panel is the Stage I video (240p) and the right panel is the Stage II video (1080p).

This model will also be very useful for enhancing and upscaling videos generated via other pipelines (e.g. HunyuanVideo).

### Open source status

- [x] The model implementation is available.
- [x] The model weights are available (Only relevant if addition is not a scheduler).

### Provide useful links for the implementation

Code: https://github.com/FoundationVision/FlashVideo/tree/main

Weights: https://huggingface.co/FoundationVision/FlashVideo