Consider support for ByteDance Bernini‑Diffusers‑v2

### Feature Idea

Hi,
I’m opening this issue to ask whether ComfyUI could consider future support for the newly released ByteDance Bernini‑Diffusers‑v2, published on HuggingFace on 2026‑08‑13.
This release includes the full Bernini pipeline:

- Qwen2.5‑VL‑7B semantic planner
- Bernini latent‑planning weights
- Wan2.2‑T2V‑A14B renderer
- SA‑3D RoPE
- full diffusers-format directory

Model:
https://huggingface.co/ByteDance/Bernini-Diffusers-v2
(official full planner+renderer release)

This version is a major upgrade over Bernini‑R and is designed for complex video generation and editing tasks that benefit from multi-step semantic planning.
I’m not requesting an immediate implementation, but I wanted to open a discussion about the possibility of supporting this new version in the future, since it is currently not integrated anywhere in the ComfyUI ecosystem.

Thanks for considering this, and thanks for all your work on ComfyUI.

### Existing Solutions

_No response_

### Other

_No response_

Start by reviewing the ByteDance/Bernini-Diffusers-v2 model page linked in the issue and its full planner-and-renderer pipeline. The issue provides no ComfyUI files, tests, or entry points; done would require an agreed implementation scope and confirmed support for the listed components.

Written by the indexing model from the issue text.