Known failing on solidcore after the smlib merge: SMLib 5e8c9c1's new blend (solidcore#2273) and main's new tests

Rig (fillets/KNOWN_FAILING.solidcore): the two 851 rows develop now rounds with their tangent
chain fail 1001 as on SMLib; the pipe on a plate's curvature round is graded to 5e8c9c1's
section, which solidcore has yet to port. Tests (KNOWN_FAILING_TESTS): main's box STL and
boss-torus tests, as on SMLib; the box-corner test and four C++ comparisons that expect
5e8c9c1's Ball corners and rounds.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01SdcgEu4Lr3moCCisdZShKz

tests/KNOWN_FAILING_TESTS

@@ -8,6 +8,8 @@
 solidcraft-cli::action_matrix::action_matrix_sample 3 baseline actions fail, the same 3 as on C++: 2 shells (Bad Loop Orient on issue-616-lid-ribs; status 1001 on filleted-shelled-box), 1 fillet (SmSurfOfRevolution Test #7 on oblique_cut_of_cylinder)
 solidcraft-cli::naming_corpus::corpus_designs_name_alike_every_build_and_fail_where_a_reference_is_gone oracle/83-boundary-loop-thicken does not load: step 8, thicken status 1001, as on C++
 solidcraft-engine::cmd::edit::tests::inserting_variable_fillet_control_preserves_parameters_at_existing_positions solidcore refuses a variable fillet whose law has a corner (solidcore#430)
+solidcraft-engine::cmd::file::tests::each_format_has_an_export_command_of_its_own the box's planar faces are meshed into 124 triangles, not 12 (the STL is 6284 bytes, the test counts 84 + 50 x 12), as on SMLib
+solidcraft-engine::measure_centre_tests::a_fillet_round_a_boss_reports_its_radius the round at the foot of a boss is not an analytic torus of minor radius 3, so no face of the body is (the test finds none to measure), as on SMLib
 solidcraft-engine::tests::chord_and_variable_fillets its 2 → 4 → 2 waist: solidcore refuses a variable fillet whose law has a corner (solidcore#430)
 solidcraft-engine::tests::fillet_fit_taper_closes_to_a_point fillet status 1001 on the taper fit (radius 16 closing to 0), as on C++
 solidcraft-kernel::blend::corner_tests::a_round_ending_in_an_earlier_round_matches_both_together fillet status 1001 on the round that ends in the earlier one, as on C++
@@ -18,5 +20,10 @@ solidcraft-kernel::blend_variable::tests::variable_radius_with_a_corner_through_
 solidcraft-kernel::corner_g2::tests::a_long_setback_keeps_the_g2_patch_inside_the_box at set backs 11 to 15 the G2 corner patch bulges past the box, as on C++
 solidcraft-kernel::corner_g2::tests::the_box_corner_g2_patch_is_a_closed_solid_curvature_continuous_to_its_neighbours the fitted G2 corner patch bulges 1.7e-5 past the box, as on C++
 solidcraft-kernel::tests::fillet_tabbed_plate_outer_loop_and_partial_selection_after_hole the R25 rim and a corner arc round together since solidcore#2223, but the two R2 corner arcs picked together fail 1001 in trim_at_creases (cliff_overflow with the crease trims off), though each builds alone (solidcore#2314)
+solidcraft-kernel::tests::fillet_three_edges_at_a_corner solidcore builds the corner as an analytic sphere octant (sphere: 1); since 38385ab5d the test wants SMLib 5e8c9c1's blend (Ball corners as rational patches, the round as its definition, smlib#79), which solidcore ports in solidcore#2273: one rational patch on the ball's sphere
+solidcraft-smlib::core::blend_tests::fillet_all_edges::solidcore_matches_the_cpp_kernel the C++ comparison: solidcore's 8 box corners are analytic spheres, C++ 5e8c9c1's rational splines (SMLib 5e8c9c1's blend (Ball corners as rational patches, the round as its definition, smlib#79), which solidcore ports in solidcore#2273)
+solidcraft-smlib::core::blend_tests::fillet_variable_linear::solidcore_matches_the_cpp_kernel the C++ comparison: solidcore 5972.141093 against C++ 5e8c9c1's 5972.142573 (SMLib 5e8c9c1's blend (Ball corners as rational patches, the round as its definition, smlib#79), which solidcore ports in solidcore#2273)
+solidcraft-smlib::core::blend_tests::rounds_on_the_871_part the C++ comparison: solidcore 3874.348250 against C++ 5e8c9c1's 3874.348458 (SMLib 5e8c9c1's blend (Ball corners as rational patches, the round as its definition, smlib#79), which solidcore ports in solidcore#2273)
+solidcraft-smlib::core::convert_tests::cpp_round_trip::a_shelled_rounded_box the C++ round trip: the shelled rounded box C++ 5e8c9c1 builds differs from solidcore's in its arrays (real array 154[0]: -10 against 0), its corners built by SMLib 5e8c9c1's blend (Ball corners as rational patches, the round as its definition, smlib#79), which solidcore ports in solidcore#2273
 solidcraft-smlib::core::offset_tests::mitred_rounds_follow_a_moved_top the C++ comparison: solidcore's offset edges keep their tolerances (offset_edges_keep_source_tolerance, solidcore#2111), C++ at 7ac037e's do not, so the moved tops' volumes are 2e-9 to 2.3e-7 apart; both are 1.2e-8 to 3.1e-7 off the closed form at accuracy 1e-8, solidcore with the fix 2.1e-7 at most
 solidcraft-smlib::ops::tests::a_variable_round_with_a_corner_in_its_law_is_its_balls_union solidcore refuses a variable fillet whose law has a corner (solidcore#430)

tests/kernel-cases/fillets/KNOWN_FAILING.solidcore

@@ -9,6 +9,8 @@
 2893-chamfer-two-distances-elliptical-rim [fail] refused, but not because "not supported yet": step 3 (solid.chamfer): SMLib chamfer: operation: SMLib failed (status 1001) (tangent propagation blended the picked edges' tangent chain with them; with it off, tangent_chain: 
 4782-fillet-bar-edge-wider-than-both-faces [fail] continuity_order: got Box: no null seams want >= g1 at every blend seam
 5158-fillet-round-just-past-a-10mm-face [fail] step 2 (solid.fillet): SMLib fillet: operation: SMLib failed (status 1001)
+851-fillet-arc-then-top-front [fail] fillet status 1001, as on SMLib 5e8c9c1 (see KNOWN_FAILING): with tangent_chain true since 21f00a917 the top front edge's chain takes the arc round's end curve, where the R2 ball sits on that round's centre circle (a torus of tube-centre radius 0, the three-edge rolling-ball corner)
+851-fillet-top-front-then-upright [fail] fillet status 1001, as on SMLib 5e8c9c1 (see KNOWN_FAILING): with tangent_chain true since 21f00a917 the upright's chain takes the first round's end curve, where the R2 ball sits on that round's axis (a torus of tube-centre radius 0, the three-edge rolling-ball corner)
 9-fillet-corner-curvature-with-reduce-fit [unsupported] solid.fillet builds `corner: g2` with `fit: exact` only
 9-fillet-corner-curvature-with-taper-fit [unsupported] solid.fillet builds `corner: g2` with `fit: exact` only
 cov-fillet-g2-corner-beside-a-cylinder [fail] step 6 (solid.fillet): the open setback corner is not a closed solid, as on SMLib d01d7b0
@@ -17,6 +19,7 @@ cov-fillet-g2-four-edge-apex-corner [fail] step 11 (solid.fillet): invalid input
 cov-fillet-g2-mixed-corner [unsupported] refused as on SMLib d01d7b0: a G2 corner is set back only where every edge meeting it is rounded and all are convex or all concave
 cov-fillet-g2-oblique-three-edge-corner [unsupported] refused on solidcore: its three edges meet at different wedge angles, and one curvature fillet takes the first edge's equal-area thumbweight for every edge (solidcore#2250, issuecomment-6563419897); SMLib d01d7b0 fails it too (see KNOWN_FAILING)
 cov-fillet-g2-pocket-corner [fail] continuity_order: G1 at one seam where the corner patch meets a contact face (g2 wanted), as on SMLib d01d7b0 (see KNOWN_FAILING)
+cov-fillet-section-curvature-pipe-on-plate-r2 [fail] solidcore builds the order-2 section as SMLib d01d7b0 did (40302.5908, G2 on both seams); the case now grades SMLib 5e8c9c1's section, the round as its definition, G3 on the pipe and 40307.8856 (fork 4a10ed8 and 79db628, to be ported in solidcore#2273)
 cov-fillet-turn-swept-boss-base-r1 [fail] as on SMLib: solid.fillet has no `turn`; the corners are mitred (12508.9675, the Mitre case's volume)
 cov-fillet-turn-swept-pocket-sharp-corner-refused-r1 [fail] as on SMLib: solid.fillet has no `turn`, so the sharp concave corners are mitred, not refused
 cov-junction-boss-tilt-20-root-round [fail] fillet status 1001 on a cylinder tilted 20 degrees into the plate, as on SMLib