fix(folderparser): use csv.reader to handle RFC 4180 quoted fields in annotation CSVs

[AbdulrahmanXAmer]

Problem

_parseAnnotationCSV used line.split(",") to extract filenames. This silently corrupts file_name when a filename contains commas inside quoted fields — valid per RFC 4180. Any dataset with such filenames fails to link images to their annotations.

Fix

Replace manual split with csv.reader (stdlib). Raw line text preserved verbatim for server upload reconstruction. File opened in text mode so \r\n normalises to \n (backward compat with Windows CSVs unchanged).

Tests

• test_parse_csv_quoted_filename — regular CSV with comma-containing filename
• test_parse_multilabel_csv_quoted_filename — _classes.csv with comma-containing filename

🤖 Generated with Claude Code

[CLAassistant]

Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you all sign our Contributor License Agreement before we can accept your contribution.
0 out of 2 committers have signed the CLA.

❌ Abdulrahman Amer
❌ pre-commit-ci[bot]

---

Abdulrahman Amer seems not to be a GitHub user. You need a GitHub account to be able to sign the CLA. If you have already a GitHub account, please add the email address used for this commit to your account.
_{You have signed the CLA already but the status is still pending? Let us recheck it.}