added sanitize to SparseCSR and __setitem__

[zerothi]

This should now work more succintly by
using a good sanitizer.
It slows down the code, a tad. The test suite
is new 3:12, vs 3:10 (with one more test).
So very little.

It should fix problem #650 and allow more easier
access to the matrix elements.

• Closes sisl.physics.hamiltonian support slice operation or not? #650
• Added tests for new/changed functions?
• Ran isort . and black . at top-level
• Changes documented in CHANGELOG.md

[codecov]

Codecov Report

Attention: 19 lines in your changes are missing coverage. Please review.

Comparison is base (e01422e) 87.65% compared to head (1a3b788) 87.64%.

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #652      +/-   ##
==========================================
- Coverage   87.65%   87.64%   -0.01%     
==========================================
  Files         364      364              
  Lines       48520    48571      +51     
==========================================
+ Hits        42532    42572      +40     
- Misses       5988     5999      +11     

Files	Coverage Δ
src/sisl/io/orca/tests/test_stdout.py	100.00% <ø> (ø)
src/sisl/sparse_geometry.py	90.30% <100.00%> (+0.01%)	⬆️
src/sisl/tests/test_sparse_geometry.py	100.00% <100.00%> (ø)
src/sisl/geometry.py	86.31% <80.00%> (-0.07%)	⬇️
src/sisl/sparse.py	93.10% <68.08%> (-1.32%)	⬇️

... and 5 files with indirect coverage changes

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