Improve SHA1 link detection

[silverwind]

This improves the SHA1 link detection to not pick up extraneous non-whitespace characters at the end of the URL. The '.' is a special case handled in code itself because of missing regexp lookahead support.

Regex test cases: https://regex101.com/r/xUMlqh/3/

Before

After

[codecov-io]

Codecov Report

Merging #6526 into master will decrease coverage by 0.02%.
The diff coverage is 69.23%.

@@            Coverage Diff             @@
##           master    #6526      +/-   ##
==========================================
- Coverage   40.28%   40.26%   -0.03%     
==========================================
  Files         403      403              
  Lines       54069    54079      +10     
==========================================
- Hits        21784    21775       -9     
- Misses      29268    29286      +18     
- Partials     3017     3018       +1

Impacted Files	Coverage Δ
modules/markup/html.go	81.93% <69.23%> (-1.37%)	⬇️
modules/avatar/avatar.go	81.25% <0%> (-18.75%)	⬇️
models/unit.go	0% <0%> (-14.29%)	⬇️
models/repo_list.go	66.84% <0%> (-1.06%)	⬇️
routers/repo/view.go	41.08% <0%> (-1%)	⬇️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 0bdd81d...3e400a2. Read the comment docs.

[silverwind]

Just in case anyone is wondering: My motiviation for this change is that I like to post fixed in <paste link to commit>. to close an issue.

The dot at the end caused the link to become invalid because the old parser appended /. to it, thinking it is a path. The new mechanism treats the dot as not part of the URL, like GitHub also does.