Support pattern-based exclusion in return_linter()

[MichaelChirico]

Closes #2335.

Implementation is maybe not ideal since it requires running the //FUNCTION | //OP-LAMBDA XPath twice -- once to find the names of assigned functions, a second time to check for lints.

The workaround I could think of was to (1) find all //FUNCTION | //OP-LAMBDA nodes (2) partition into assigned vs. unassigned (3) trim except|except_regex from the assigned group (4) combine_nodesets() (4) check for lints. But IINM this would require refactoring a lot of other code as well, I'm not sure it's worth it -- //FUNCTION | //OP-LAMBDA should be relatively fast.

[codecov-commenter]

Codecov Report

All modified and coverable lines are covered by tests ✅

Comparison is base (c912988) 98.53% compared to head (210bfaa) 98.53%.

❗ Current head 210bfaa differs from pull request most recent head 586ab3b. Consider uploading reports for the commit 586ab3b to get more accurate results

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #2433   +/-   ##
=======================================
  Coverage   98.53%   98.53%           
=======================================
  Files         126      126           
  Lines        5659     5670   +11     
=======================================
+ Hits         5576     5587   +11     
  Misses         83       83           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[AshesITR]

Very nice feature - makes the linter much more attractive for larger code bases.