Feat/gitlab adapter content

[jaysomani]

No description provided.

[chatgpt-codex-connector]

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.

[greptile-apps]

Greptile Summary

This PR implements getRepositoryTree, getRepositoryContent, and listRepositoryContents for the GitLab adapter, replacing previous Not implemented stubs. Previous review concerns (recursive pagination, hardcoded main ref, unvalidated base64_decode) are all addressed, but two new pagination gaps remain:

• The non-recursive getRepositoryTree path makes a single request without per_page, silently truncating at GitLab's default of 20 root-level items.
• listRepositoryContents has the same issue — no per_page and no pagination loop, so directories with more than 20 entries will return incomplete results.

Confidence Score: 3/5

Two P1 pagination bugs will silently return incomplete data for any repo or directory with more than 20 items; should be fixed before merging.

The recursive tree pagination loop (the previous P1 concern) is properly fixed, and the ref/base64 issues are also resolved. However, the non-recursive tree path and listRepositoryContents both omit per_page and pagination, which will cause silent data truncation in production on repos beyond GitLab's 20-item default — a present correctness defect on the changed path.

src/VCS/Adapter/Git/GitLab.php — non-recursive getRepositoryTree (line 206) and listRepositoryContents (line 261) both need per_page=100 and pagination loops.

Important Files Changed

Filename	Overview
src/VCS/Adapter/Git/GitLab.php	Implements getRepositoryTree (with pagination for recursive mode), getRepositoryContent (using HEAD ref fallback, validated base64 decode), and listRepositoryContents — but both the non-recursive tree path and listRepositoryContents lack per_page/pagination, silently truncating results at GitLab's default of 20 items.
tests/VCS/Adapter/GitLabTest.php	Adds integration tests for all newly implemented methods; tests use small datasets (≤3 items) so they pass today but won't catch the pagination truncation in production scenarios.

_{Reviews (2): Last reviewed commit: "updated with suggestions" | Re-trigger Greptile}