Fix isort failure in pre-commit hooks

[WeilerP]

IMPORTANT: Please search among the Pull requests before creating one.

Description

• Bumps isort to 5.12.0.

Closes

Closes #993.

[codecov]

Codecov Report

Merging #994 (1f697fe) into master (525b847) will decrease coverage by 1.39%.
The diff coverage is n/a.

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #994      +/-   ##
==========================================
- Coverage   82.19%   80.81%   -1.39%     
==========================================
  Files          58       58              
  Lines        8763     8763              
  Branches     1653     1653              
==========================================
- Hits         7203     7082     -121     
- Misses       1028     1155     +127     
+ Partials      532      526       -6     

Impacted Files	Coverage Δ
cellrank/_utils/_lineage.py	83.83% <ø> (ø)
...timators/terminal_states/_term_states_estimator.py	84.00% <ø> (ø)
cellrank/models/_base_model.py	86.09% <ø> (ø)
cellrank/models/_sklearn_model.py	84.21% <ø> (ø)
cellrank/_utils/_linear_solver.py	41.20% <0.00%> (-52.20%)	⬇️
cellrank/datasets/_datasets.py	47.76% <0.00%> (-17.92%)	⬇️
cellrank/external/kernels/_wot_kernel.py	84.74% <0.00%> (-3.39%)	⬇️
cellrank/kernels/_utils.py	68.46% <0.00%> (-2.71%)	⬇️
cellrank/_utils/_parallelize.py	75.67% <0.00%> (-2.71%)	⬇️
cellrank/estimators/mixins/decomposition/_schur.py	92.50% <0.00%> (-2.50%)	⬇️
... and 1 more

[WeilerP]

Not sure why but the pre-commits passed on a macOS machine but stripped unnecessary # noqa's on a Linux machine.

[WeilerP]

For future reference: CI failing is caused by some issue with codecov.