Add support to Meissonic

Meissonic is a non-autoregressive mask image modeling text-to-image synthesis model that can generate high-resolution images. It is designed to run on consumer graphics cards.

The model checkpoint can be found in https://huggingface.co/MeissonFlow/Meissonic
The inference code can be found in https://github.com/viiika/Meissonic
The paper can be found in https://arxiv.org/abs/2410.08261

Anyone interested in implementing this?

Start with the Meissonic checkpoint, linked inference code, and paper to understand the model and expected behavior. Then identify the integration surface in diffusers and validate that the added support reproduces the linked inference results; done means Meissonic can generate high-resolution images on consumer graphics cards.

Vom Indexierungsmodell aus dem Issue-Text verfasst.