fix: handle scalar string groups in ranked-gene plots

[yingluosanqian]

Description

groups accepts either a single string or a sequence of strings in the ranked-gene plotting APIs. However, sc.pl.rank_genes_groups and the shared plotting path iterated a scalar string character by character.

For example, passing groups="Dendritic" attempted to look up the group "D" and raised an error instead of plotting "Dendritic".

This PR normalizes scalar string group names to one-element lists before iteration, consistent with the existing behavior in sc.pl.rank_genes_groups_violin and sc.get.rank_genes_groups_df.

• Closes # — no existing issue
• Tests included: added regression coverage for both the standalone and shared plotting paths

• Release notes not necessary because: it's just a simple fix.

Tests

• add related test.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 81.99%. Comparing base (a656a33) to head (34b490b).
✅ All tests successful. No failed tests found.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #4330   +/-   ##
=======================================
  Coverage   81.98%   81.99%           
=======================================
  Files         134      134           
  Lines       13235    13239    +4     
=======================================
+ Hits        10851    10855    +4     
  Misses       2384     2384           

Flag	Coverage Δ
hatch-test.low-vers	79.07% <100.00%> (+<0.01%)	⬆️
hatch-test.pre	81.86% <100.00%> (+<0.01%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Files with missing lines	Coverage Δ
src/scanpy/plotting/legacy/_tools/__init__.py	76.17% <100.00%> (+0.25%)	⬆️