Modernize pyExtractHEIC: proper package structure, CLI, tests, and CI

[Copilot]

The original codebase was a single script with several correctness bugs, hardcoded broken paths, no tests, and no CI. This PR refactors it into a maintainable Python package.

Bug fixes

• Inverted extraction logic — if not fnmatch(…, "icloud photos/*") extracted files outside the iCloud folder; flipped to if fnmatch(…)
• Silent cleanup failure — os.rmdir() raises on non-empty directories; replaced with shutil.rmtree()
• Uppercase .HEIC skipped — case-sensitive fnmatch("*.heic") missed iPhone files; lowercased before matching
• Broken paths — r"\zip" resolved to /zip (filesystem root); replaced with Path(__file__).parent / "zip"

Package restructure

• heic_converter/converter.py — decomposed into three independently testable functions: extract_heic_from_zip, convert_heic_to_png, convert_directory, and an orchestrator process_zip_dir
• heic_converter/cli.py — argparse CLI with --zip-dir, --extract-dir, --convert-dir, --zip-pattern, -v; invocable as python -m heic_converter
• ConvertIphonePics.py — retained as a thin shim delegating to the package for backward compatibility

Tooling

• pyproject.toml — packaging, ruff, and pytest config
• requirements.txt — pip install -r requirements.txt works out of the box
• .github/workflows/ci.yml — ruff + pytest on every push/PR with permissions: contents: read
• .gitignore — excludes zip files, artifacts, and the previously committed iCloud Photos (11).zip binary (removed from working tree)

Tests

13 unit tests covering extraction, conversion, end-to-end orchestration, and CLI error handling — including case-insensitive matching, missing directories, empty zips, and non-matching zip patterns.

---

💬 We'd love your input! Share your thoughts on Copilot coding agent in our 2 minute survey.