Size the band mask on the mesh the labels are read from

place_fault_ribbon_2d built the per-strand footprint mask from an array
sized on the mesh BEFORE add_fault, then filled it from labels read on
the mesh after. The two agree whenever the cut adds no cells, which is
the usual case with embedded spines, and disagree as soon as it does:
the coarse rig with a short junction ligament refused with a broadcast
error, 2582 cells against 2562.

Underworld development team with AI support from Claude Code

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RzK7JXSxsc2GoG7TGnQgUS

Author: lmoresi

src/underworld3/utilities/place_surface.py

@@ -8950,7 +8950,11 @@ def place_fault_ribbon_2d(base_mesh, traces, width, *, margin_rings=2,
     # belong to no trace). A band cell belongs to the trace whose USER
     # polyline is nearest, within the band's reach of it and not past
     # its ends by more than half a rung (_footprints_by_user_polyline).
-    band = np.zeros_like(band_all)
+    # sized on the CURRENT mesh, not on band_all: a cut adds cells where
+    # the trace does not already lie on mesh edges, so after add_fault the
+    # chart is longer than the one band_all was built for (measured: 2582
+    # against 2562 on the coarse rig with a short junction ligament)
+    band = np.zeros(int(mesh.dm.getHeightStratum(0)[1]), dtype=bool)
     for k in range(len(spines)):
         band |= mesh.cells_labelled(band_label, band_value + k)
     S_all = np.vstack(extended)