perf: skip is_package() network call for already-synced Drive files

[Copilot]

• Identify root cause: is_package() in drive_parallel_download.py makes a network request for every file before checking local existence
• Fix collect_file_for_download in drive_parallel_download.py to check local file existence FIRST before making any network request
• Fix process_file in sync_drive.py (legacy function) with same optimization, eliminating duplicate package post-download code via if/elif/else structure
• Improve comments to clearly explain the side-effect of package_exists() and the directory-implies-package assumption
• Add test test_collect_file_for_download_package_outdated to cover the new early-return path for outdated package directories
• Identify CI coverage failure: drive_file_existence.py lines 68-69 lost coverage because the new code only calls package_exists() when os.path.isdir() is True (old code called it for first-time package downloads with non-existent paths)
• Add test_package_exists_none_item and test_package_exists_non_existent_path to restore 100% coverage of drive_file_existence.py
• All 92 drive sync tests pass; all changed files at 100% coverage; linter clean

[Copilot]

Pull request overview

This PR improves iCloud Drive sync performance by avoiding the expensive is_package() HTTP request for items that are already present locally and up-to-date. It updates both the parallel download path and the legacy process_file() path to check local existence first, and adds targeted tests to keep coverage at 100%.

Changes:

• Reordered Drive sync decision logic to check os.path.isfile() / os.path.isdir() and local freshness (file_exists() / package_exists()) before calling is_package().
• Added an early-return download-task path for “outdated local package directory” cases to re-download packages without an extra is_package() network call.
• Added/expanded unit tests to cover package_exists() edge cases and the new early-return behavior.

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated no comments.

File	Description
src/drive_parallel_download.py	Skips is_package() for already-synced files/packages by checking local existence first; adds fast re-download scheduling for outdated package dirs.
src/sync_drive.py	Applies the same optimization to legacy process_file() and simplifies the control flow with an explicit if/elif/else structure.
tests/test_sync_drive.py	Adds tests covering package_exists() falsey cases and the new “outdated package dir” early-return path in collect_file_for_download.