Add Lumina-DiMOO as a pipeline

### Model/Pipeline/Scheduler description

Lumina-DiMOO is a unified multimodal model built on fully discrete diffusion. It supports a wide range of tasks, including text-to-image, image-to-image (editing, subject-driven generation, inpainting), and image understanding. It also outperforms existing open-source multimodal models with stronger results and much higher sampling efficiency. It would be great to have this model in diffusers.

### 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

project page: https://synbol.github.io/Lumina-DiMOO
code: https://github.com/Alpha-VLLM/Lumina-DiMOO
model weights: https://huggingface.co/Alpha-VLLM/Lumina-DiMOO

@synbol @ChinChyi

Start by reading the linked Lumina-DiMOO project page and implementation, then inspect the model weights on Hugging Face. Determine how its text-to-image, image-to-image, inpainting, subject-driven generation, and image-understanding tasks map to a diffusers pipeline. Done means Lumina-DiMOO is available in diffusers with the requested model capabilities.

Written by the indexing model from the issue text.