fix(image): flatten transparent images onto white instead of black

[wfatih]

ImageDocumentBackend converts every frame with frame.convert("RGB"). That drops the alpha channel and keeps whatever colour is stored under it, and for fully transparent pixels that is usually black. A PNG with dark text on a transparent background (RGBA, LA, or palette + tRNS) therefore becomes an all-black page, and the text is invisible to layout and OCR:

img = Image.new("RGBA", (64, 48), (0, 0, 0, 0))   # transparent background
img.paste((0, 0, 0, 255), (0, 0, 16, 12))          # black "text"
# get_page_image().getpixel((63, 47)) -> (0, 0, 0) on main

This adds _to_rgb(), which composites frames that have an alpha channel or a transparency entry onto a white background before the RGB conversion, the same white a PDF page has behind its content. Frames without transparency still go through a plain convert("RGB"), so their output doesn't change.

The new test, parametrised over RGBA, LA and P + tRNS, checks that the transparent background comes out white and the opaque content stays black. It fails on main (3 failed) and passes with this change. tests/test_backend_image_native.py (30 passed) and tests/test_input_doc.py (24 passed) pass, and no groundtruth changes: the only RGBA test image (a TIFF) is fully opaque.

Checklist:

• Documentation has been updated, if necessary.
• Examples have been added, if necessary.
• Tests have been added, if necessary.

[mergify]

Merge Protections

🟢 Merge protection satisfied — ready to merge.

Show 1 satisfied protection

🟢 Enforce conventional commit

Make sure that we follow https://www.conventionalcommits.org/en/v1.0.0/

• title ~= ^(fix|feat|docs|style|refactor|perf|test|build|ci|chore|revert)(?:\(.+\))?(!)?:

[github-actions]

✅ DCO Check Passed

Thanks @wfatih, all your commits are properly signed off. 🎉