update solver to default dtype float64

[m-reuter]

Solver was sometimes taking dtype from the mesh (usually float32) and sometimes forcing it to float32 in tria_aniso. This was leading to issues (see #104 ).

We now always allow a dtype argument in solver and default to float64, which will require more memory but will give higher precision. This should be similar to other FEM packages and scipy etc.

Users who run into memory issues should pass dtype=float32 explicitly.

We may break backward compatibility here, by switching to float64 which will change values (should be more accurate now).

Speed should not be affected much.

[Copilot]

Pull request overview

This PR addresses issue #104 by standardizing the dtype handling in the Solver class. Previously, the anisotropic FEM method was forcing matrices to float32 while the isotropic method used the mesh's dtype (typically float32). This inconsistency caused precision issues in downstream computations.

Changes:

• Added a dtype parameter to the Solver class and all FEM computation methods, defaulting to np.float64 for improved numerical stability
• Updated all matrix construction code to respect the specified dtype through proper conversion using astype(dtype, copy=False)
• Updated test expectations for test_normalize_ev_geometry to reflect the improved precision from float64

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 4 comments.

File	Description
lapy/solver.py	Added dtype parameter throughout the Solver class hierarchy; updated init, _fem_tria, _fem_tria_aniso, fem_tria_mass, _fem_tetra, and _fem_voxels methods to accept and use dtype parameter; added dtype conversion logic to all matrix construction code; improved docstring grammar and clarity
lapy/utils/tests/expected_outcomes.json	Updated expected values for test_normalize_ev_geometry test to reflect improved precision from float64 default

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[codecov]

Codecov Report

❌ Patch coverage is 46.66667% with 16 lines in your changes missing coverage. Please review.
✅ Project coverage is 44.41%. Comparing base (ab3cc22) to head (643507d).
⚠️ Report is 4 commits behind head on main.

Files with missing lines	Patch %	Lines
lapy/solver.py	46.66%	16 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #105      +/-   ##
==========================================
- Coverage   44.42%   44.41%   -0.01%     
==========================================
  Files          15       15              
  Lines        2753     2758       +5     
  Branches      348      348              
==========================================
+ Hits         1223     1225       +2     
- Misses       1408     1411       +3     
  Partials      122      122              

Flag	Coverage Δ
unittests	44.41% <46.66%> (-0.01%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ 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.