test: cover image transform helpers used by Idefics3 and Jina CLIP preprocessing

[Aditya17-bot]

Adds unit tests for the image transform helpers that only ran through model tests (which download weights): pad2square, crop_ndarray, resize_ndarray, ResizeForVisionEncoder, ImageSplitter and SquareResize. They are used by Jina CLIP (PadtoSquare) and ColModernVBERT (Idefics3 resizing and splitting).

What the tests pin:

• pad2square: a smaller image is pasted top-left on the fill color; a larger one is center-cropped with round-half-to-even offsets (7 -> 2, 9 -> 2), matching the torchvision note in the code.
• crop_ndarray: channel-first and channel-last crops agree.
• resize_ndarray: size is (width, height), the layout stays (C, H, W), uint8 stays uint8, and float images stay in [0, 1].
• ResizeForVisionEncoder: landscape, portrait, smaller-than-one-tile and exact-tile inputs round up to multiples of the tile size, with the same sizes as resize_for_vision_encoder in transformers' Idefics3 image processor.
• ImageSplitter: small images pass through untouched; larger ones give rows * cols tiles in row order plus a 512x512 global view; per-image grouping is kept.
• SquareResize: each image is wrapped in its own list at size x size.

ImageSplitter is only tested on multiples of the tile size, since ResizeForVisionEncoder always runs first. For other sizes it splits into evenly sized patches, whereas transformers' current split_images takes fixed tiles, so I didn't want to pin that path.

The tests are in a new file so they don't conflict with #760 and #761, which both edit tests/test_image_transform.py. I left resize_longest_edge out because #761 covers it.

Coverage from tests/test_image_transform.py alone vs. with the new file:

• functional.py: 57% -> 82%
• operators.py: 43% -> 62%

$ pytest tests/test_image_transform_ops.py tests/test_image_transform.py
29 passed
$ ruff format --check / ruff check: clean

• Have you followed the guidelines in our Contributing document?
• Have you checked to ensure there aren't other open Pull Requests for the same update/change?
• Does your submission pass the existing tests?

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

• Configuration used: Organization UI
• Review profile: CHILL
• Plan: Advanced
• Run ID: 65846d75-3e03-499a-8e80-de61ced4d165

📥 Commits

Reviewing files that changed from the base of the PR and between c0588ac and 8f3072a.

📒 Files selected for processing (1)

• tests/test_image_transform_ops.py

Included review availability: This review used your included allowance. Your plan provides up to 8 included reviews per hour; 7 remain after this review.

---

📝 Walkthrough

Walkthrough

The change adds tests for image padding, cropping, and resizing, including channel layouts and output dtype expectations. It adds assertions for vision-encoder dimensions, image-splitting tile contents and counts, per-image grouping, and SquareResize frame shapes. It does not change production code.

Priority: ⬇️ Low

Estimated code review effort: 2 (Simple) | ~12 minutes

Change: Other

Merge Risk: ⚪ Minimal · up to 8f307

No merge-blocking issue is identified; merge after normal checks.

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 8.33% which is insufficient. The required threshold is 80.00%. Docstring coverage is scoped to functions touched by this diff. Analyzed 12 functions across 1 files.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title clearly summarizes the main change: adding tests for image transform helpers used in Idefics3 and Jina CLIP preprocessing.
Description check	✅ Passed	The description explains which image transform helpers the tests cover and summarizes the tested behavior and reported validation results.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

• Fix all pre-merge checks with AI

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create a new PR

• Autopilot · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}