ltfs_ordered_copy QoL features (type annotations, etc.)

まず ltfs_ordered_copy スクリプトを見つけ、リポジトリがサポートしている Python のバージョンを確認します。提案されている代替案（stub ファイル、docstring、またはより新しい言語機能）を検討し、作業内容を定義する前に、maintainer の互換性に関する方針を確認します。合意され、範囲を限定したモダナイズの作業が、指定されていない互換性変更なしに実装されていれば完了とします。

Is your feature request related to a problem? Please describe.
Not a problem per se, rather a minor inconvenience (and offer for contribution): the ltfs_ordered_copy script lacks several "creature comforts"/conveniences found in modern python modules (e.g.: type annotations) - I assume this is with a view to maximising compatibility.

Describe the solution you'd like
Without knowing the appetite of the maintainers for breaking compatibility with older python versions (and/or introducing an entirely new script with greater minimum python version requirements - messy IMHO), the following solutions could be considered:

• (non-breaking) a low-friction solution for type annotations specifically would be a stub file.
• (non-breaking) adding docstrings (optionally with an established style, e.g.: numpy) aids readability.
• (breaking) if a (new) minimum version can be established, various other readability improvements enabled by in-language features (e.g.: f-strings, more concise/readable variants of standard library functions (such as those in pathlib)) could be implemented.

Describe alternatives you've considered
The above are three "alternatives" (non-mutually-exclusive) I've considered; a fourth would be "do nothing". I personally plan to utilise some of the classes within my own work, so will likely invest effort "modernising" regardless - while I could hypothetically maintain a fork of this entire repository for the sake of one script, this overhead is a disincentive, and realistically other users are unlikely to stumble across such a fork; including modernising updates in the upstream makes this work available to a broader user base.