Resample all LMS cube planes in one step

[astronomyk]

What

The FieldOfView branch of MetisLMSSpectralTraceList.apply_to built a fresh RectBivariateSpline(kx=1, ky=1) over the field-of-view plane for every wavelength plane of the cube and evaluated it at the slice's (yfov, xfov) coordinates. Those sample coordinates depend only on the slice geometry — they are identical for every plane — so with ~2,500–7,000 planes (depending on !SIM.spectral.spectral_bin_width) and 28 slices this meant up to ~200,000 spline constructions per observation. In a profiled LMS run this loop accounted for the bulk of observe().

The new module function interpolate_cube_planes(fovcube, yfov, xfov) computes the bilinear indices and weights once per slice and gathers all planes in a single vectorised expression. Coordinates outside the plane are clamped to the edges, exactly as RectBivariateSpline does.

End-to-end check (full lms mode, point source, identical inputs, separate processes): observe() wall time 386.5 s -> 108.6 s, with the detector image identical to the last digit (sum 1617362510.5459003, max 193.6031494140625 on both). The like-for-like loop comparison on one slice of the real cube is 10x (2.89 s -> 0.29 s, x28 slices per observation).

Testing

• TestInterpolateCubePlanes::test_matches_spline_per_plane_reference — exact equivalence against the per-plane spline evaluation on random data, including sample coordinates beyond the plane edges (clamping behaviour).
• A notebook with the full LMS observation and the like-for-like timing on the real cube is attached below.

[codecov]

Codecov Report

❌ Patch coverage is 90.00000% with 1 line in your changes missing coverage. Please review.
✅ Project coverage is 76.51%. Comparing base (86ad453) to head (8ffc9c9).

Files with missing lines	Patch %	Lines
scopesim/effects/metis_lms_trace_list.py	90.00%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #995      +/-   ##
==========================================
+ Coverage   76.47%   76.51%   +0.04%     
==========================================
  Files          69       69              
  Lines        9027     9032       +5     
==========================================
+ Hits         6903     6911       +8     
+ Misses       2124     2121       -3     

☔ View full report in Codecov by Harness.
📢 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.