[nettanvirdev]

What does this PR do?

load_video() documents that frames are converted to "RGB" when convert_method=None, but it returned them unchanged. For GIFs, Pillow decodes the first frame as P and later frames as RGB/RGBA, so the mixed modes made VideoProcessor.preprocess_video() and AnimateDiffVideoToVideoPipeline fail with ValueError: all input arrays must have the same shape.

This adds the missing else branch so every frame is converted with image.convert("RGB"), the same as load_image. A custom convert_method is still applied as before and skips the conversion.

Fixes #15098

Notes:

• The conversion happens once after both decode paths. For mp4/imageio frames it is a no-op on the pixels (they are already RGB arrays); I checked a locally written mp4 and the frames are identical to the raw imageio frames.
• Transparent GIFs (P then RGBA frames) now come back as RGB, matching what load_image returns for the same frames.
• No "Copied from" counterpart of load_video exists, so nothing else needs updating. No docs change needed; the docstring already describes this behavior.

Tests

Added TestLoadVideo::test_gif_frames_are_rgb in tests/others/test_utils.py.

$ pytest tests/others/test_utils.py -k TestLoadVideo
# on main
E         At index 0 diff: 'P' != 'RGB'
FAILED tests/others/test_utils.py::TestLoadVideo::test_gif_frames_are_rgb
# with this PR
1 passed, 26 deselected

$ pytest tests/others/test_utils.py tests/others/test_video_processor.py
1 failed, 33 passed, 1 skipped

The one failure is TestDeprecate::test_deprecate_stacklevel, which also fails on main on Windows (it asserts a /-separated path against a \-separated filename).

Also ran ruff check / ruff format --check (0.9.10), utils/check_copies.py, utils/check_dummies.py, utils/check_doc_toc.py and doc-builder style --check_only: all clean.

Before submitting

• Did you use an AI agent (Claude Code, Codex, Cursor, etc.) to help with this PR? If so:
  • Did you read the Coding with AI agents guide?
  • Did you run the self-review skill on the diff?
  • Did you share the final self-review notes in the PR description or a comment?
• Did you read the contributor guideline?
• Did you read our philosophy doc? (important for complex PRs)
• Was this discussed/approved via a GitHub issue or the forum? Please add a link to it if that's the case.
• Did you make sure to update the documentation with your changes? Here are the
  documentation guidelines, and
  here are tips on formatting docstrings.
• Did you write any new necessary tests?
• Are you the author (or part of the team) of the model/pipeline (only applicable for model/pipeline related PRs)?

Who can review?

@sayakpaul @DN6