feature: Reads the size and ratio of the image and gives the recommended size

[xhoxye]

[mashb1t]

@xhoxye thanks for contributing, will check this out in detail in a few hours. Please find one commend attached.

[mashb1t]

@xhoxye I've applied some adjustments, please check again

[xhoxye]

Thank you very much for your help, although it is a little different from what I had in mind at the beginning, I initially wanted to recommend the standard size of SDXL

[mashb1t]

@xhoxye thanks for the feedback, that's what i assumed in #2971 (comment) and what makes the most sense.
My reasoning was that people might train LoRAs with specific aspect ratios or bucketing with other resolutions, but when the base model isn't trained these aspect ratios aren't the best choice anyways.
Users can still decide to just choose another available aspect ratio.

[xhoxye]

I modified the use of textbox to display information to facilitate the translation of titles


desc_image_size = gr.Textbox(label='Image Size and Recommended Size', elem_id='desc_image_size', visible=False)
size_info = f'{width} x {height}, {ratio}, {lcm_ratio}' size_info += f'\n{recommended_width} x {recommended_height}, {recommended_ratio}, {recommended_lcm_ratio}'
Why gr. Markdown doesn't show label?

[mashb1t]

Will include it in a few hours. Please add a new PR to develop if you want to be the author, if not i'll just cherry pick from your branch.