Performance: use ImageData when grid has uniform axis spacing

[banesullivan]

Use pv.ImageData over pv.RectilinearGrid when the axes have uniform spacing.

This yields significant performance improvements, especially if wanting to volume render:

import xarray as xr
import pvxarray
import pyvista as pv

ds = xr.tutorial.load_dataset("cells3d")
da = ds.images
nuclei = da.sel(c='nuclei').pyvista.mesh(x="x", y="y", z="z")

pl = pv.Plotter()
pl.add_volume(nuclei, clim=(0, 30000), opacity='sigmoid')
pl.enable_terrain_style()
pl.show()

The tests here are working around an upstream bug: pyvista/pyvista#7686

maintenance

• Updates the GitHub Actions to run again
• Bump the required version of PyVista
• Fix deprecation issues
• More python testing versions

[codecov-commenter]

Codecov Report

❌ Patch coverage is 80.95238% with 12 lines in your changes missing coverage. Please review.
✅ Project coverage is 92.90%. Comparing base (cc606e4) to head (761029b).
⚠️ Report is 1 commits behind head on main.

Files with missing lines	Patch %	Lines
pvxarray/structured.py	40.00%	6 Missing ⚠️
pvxarray/rectilinear.py	84.00%	4 Missing ⚠️
pvxarray/accessor.py	86.66%	2 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main      #81      +/-   ##
==========================================
- Coverage   93.60%   92.90%   -0.70%     
==========================================
  Files          11       11              
  Lines         657      691      +34     
==========================================
+ Hits          615      642      +27     
- Misses         42       49       +7     

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

🚀 New features to boost your workflow:

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