fix: scalar setting/reshape on CoW .X

[ilan-gold]

@flying-sheep I'm ready to merge #2338 but noticed I hadn't backported this feature with the CoW branch. I'm actually not even 100% sure we should support these shortcuts. WDYT? I'll hold off on merging the other PR until I hear back.

• Closes #
• Tests added
• Release note added (or unnecessary)

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 86.89%. Comparing base (60226ed) to head (f3ada7d).
⚠️ Report is 22 commits behind head on 0.12.x.

Additional details and impacted files

@@            Coverage Diff             @@
##           0.12.x    #2340      +/-   ##
==========================================
+ Coverage   84.82%   86.89%   +2.06%     
==========================================
  Files          46       46              
  Lines        7251     7259       +8     
==========================================
+ Hits         6151     6308     +157     
+ Misses       1100      951     -149     

Files with missing lines	Coverage Δ
src/anndata/_core/anndata.py	83.80% <100.00%> (+0.16%)	⬆️

... and 7 files with indirect coverage changes

[flying-sheep]

I'm actually not even 100% sure we should support these shortcuts

Hmm, maybe only if X is not None so we don’t have to guess the container data type?

Anyway, I don’t have a strong opinion here

[ilan-gold]

Hmm, maybe only if X is not None so we don’t have to guess the container data type?

That's a good point. But then what if it's sparse? We do a dense-sparse matrix? I think I will add a warning that setting via scalars (and reshaping) will be deprecated at some point and for now it is np.full when scalar