fix: strip trailing CJK/ASCII punctuation from bare URLs

[tiensonqin]

Summary

Bare autolinks were incorrectly including trailing punctuation:

• Chinese ， / 。 (and related CJK punct) after a URL were absorbed into the link (also reported in Chinese period is parsed into URL logseq#7634)
• English , / . after host-only URLs like http://example.com. were included (path URLs already stripped these)

Root cause: ending punctuation was only handled as ASCII chars in excluded_ending_chars. Multi-byte UTF-8 CJK punctuation could never match, and the host (before_path) parser did not apply trailing-ASCII rules at all.

Fix

• Treat CJK punctuation (，。；！？、：) as hard URL terminators (never part of the URL)
• Apply ASCII trailing punctuation stripping to the host part as well as the path
• Extend string_contains_balanced_brackets with excluded_ending_strings for multi-byte stoppers

CI

• Enable dune-cache: true on ocaml/setup-ocaml in Build and Format workflows (opam switch cache was already automatic)

Test plan

• Added cases for host/path + English , . and Chinese ， 。
• dune runtest passes
• ocamlformat clean

  

[CLAassistant]

Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you sign our Contributor License Agreement before we can accept your contribution.
_{You have signed the CLA already but the status is still pending? Let us recheck it.}

[Copilot]

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.