[Issue]: img2img Image tab edit button controls are confusing and unclear what they actually supposed to do

[AznamirWoW]

Issue Description

Open img2img. On the image tab load an image of a fox face, provide a prompt "a red fox".
Click Edit button, see this:

1. mouse scroll resizes the uploaded image, but it has no effect on the result
   

2. left mouse click revert the image size to original, but clicking generate results in
   

3. another left mouse click resizes the image, now it has black borders around, but clicking generate results in
   

reload the image again, click edit - the image has a blue border around it that can be dragged either by an edge or by a corner

4. dragging bottom right corner to this position
   
   after releasing the mouse button the image and the border snaps to original

but clicking generate reveals that the actual selection is the top left quadrant, which renders as

left clicking the image reveals that is the case:

but generate again renders the original and not what is seen on the screen

Version Platform Description

08:19:53-920806 INFO Logger: file="G:\SD\sdnext.log" level=INFO size=2479074 mode=append
08:19:53-922806 INFO Python version=3.10.11 platform=Windows bin="G:\SD\venv\Scripts\python.exe" venv="G:\SD\venv"
08:19:54-037832 INFO Version: app=sd.next updated=2024-07-25 hash=42542562 branch=dev

Relevant log output

No response

Backend

Diffusers

UI

Standard

Branch

Dev

Model

StableDiffusion 1.5

Acknowledgements

• I have read the above and searched for existing issues
• I confirm that this is classified correctly and its not an extension issue

[solsol360]

Nobody seems to have any idea how this works even in Discord support threads.