Qwen image img2img pipeline not working correctly

### Custom Node Testing

- [x] I have tried disabling custom nodes and the issue persists (see [how to disable custom nodes](https://docs.comfy.org/troubleshooting/custom-node-issues#step-1%3A-test-with-all-custom-nodes-disabled) if you need help)

### Expected Behavior

When running img2img using queen image I'd expect results like those shown [here](https://github.com/huggingface/diffusers/pull/12117) with some reasonable understanding of image and style out of the box.

### Actual Behavior

Results are garbled and look overcooked such as in [here](https://github.com/comfyanonymous/ComfyUI/issues/9769)

### Steps to Reproduce

Just regular img2img pipeline with qwen image (not qwen image edit)

### Debug Logs

```powershell
since the workflow works i do not get error messages so logs are not helpful here.
```

### Other

As per the first link in the diffusers ibraries, I have the suspicion that qwen requires an slightly different pipeline for img2img and inpainting that is simply not implemented.