Fix/windows int32 issue

[d-v-b]

There's a windows-specific issue where np.dtype('i') is an instance of np.dtypes.IntDType instead of np.dtypes.Int32DType, which breaks dtype inference. This PR adds logic to the dtype inference for Int32 to handle this case, and adds a test that fails without that fix.

TODO:

• Add unit tests and/or doctests in docstrings
• Add docstrings and API docs for any new/modified user-facing classes and functions
• New/modified features documented in docs/user-guide/*.rst
• Changes documented as a new file in changes/
• GitHub Actions have all passed
• Test coverage is 100% (Codecov passes)

[d-v-b]

rtd build is failing but I can't replicate this locally...

[jhamman]

docs should be fixed now. Thanks @d-v-b

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 94.48%. Comparing base (bbdefac) to head (dacc079).
⚠️ Report is 130 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #3151   +/-   ##
=======================================
  Coverage   94.47%   94.48%           
=======================================
  Files          78       78           
  Lines        8604     8607    +3     
=======================================
+ Hits         8129     8132    +3     
  Misses        475      475           

Files with missing lines	Coverage Δ
src/zarr/core/dtype/npy/int.py	98.73% <100.00%> (+0.01%)	⬆️

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.