fix(core): handle parameter limits in files metadata deletion

[gitedmond]

Refactor dropMetadataForFiles and dropIndexForFiles to chunk by 500 and wrap in database transactions to avoid query parameter limit errors on Oracle and PostgreSQL.

Assisted-by: Antigravity:Gemini-3.5-Flash

• Resolves: [Bug]: files:scan creates oversized files_metadata DELETE query and causes severe resource usage #62325

Summary

Refactor dropMetadataForFiles and dropIndexForFiles to chunk by 500 and wrap in database transactions to avoid query parameter limit errors on Oracle and PostgreSQL.

It looks like there was already some chunking logic set up here, but it was accidentally passing the original huge file id array instead of the chunked one. This pr just swaps that variable out so chunking actually works and wraps the loop in a transaction to keep the db happy.

Changes

• Corrected variable references: Changed the query builder parameter binder inside the loops to use the chunked $chunk variable instead of the original $fileIds array.
• Set safe chunk size of 500 to prevent queries from exceeding database parameter limits.
• Wrapped loops in transaction blocks (beginTransaction / commit / rollBack) to optimize performance and prevent inconsistency on failure.
• Added unit tests verifying chunking logic and transaction handling in tests/lib/FilesMetadata/FilesMetadataManagerTest.php.

Checklist

• Code is properly formatted
• Sign-off message is added to all commits (Note: You must run git commit --amend -s on the branch to add your DCO signature)
• Tests (unit, integration, api and/or acceptance) are included
• Screenshots before/after for front-end changes
• Documentation (manuals or wiki) has been updated or is not required
• Backports requested where applicable (ex: critical bugfixes)
• Labels added where applicable (ex: bug/enhancement, 3. to review, feature component)
• Milestone added for target branch/version (ex: 32.x for stable32)

AI (if applicable)

• The content of this PR was partly or fully generated using AI