feat(ets): fkine/jacob0/jacobe/hessian0/hessiane accept compact q

[petercorke]

Summary

• ikine_LM/ik_LM and friends return a solution sized to just the joints on the requested sub-chain (e.g. YuMi's l_gripper, 7 elements) -- but fkine/jacob0/jacobe/hessian0/hessiane only understood a full, global jindex-addressed q (14 elements for YuMi), silently misindexing when handed the shorter compact solution directly. This is the root cause of Dual Arm robot (YuMi robot) gives different solutions, when check using forward kinematics, gives different trajectories each time... #379's remaining "fkine gives garbage" symptom, on top of the gripper-labelling bug already fixed in fix(models,ik): correct YuMi gripper parents and IK failure-path q compaction #649.
• Added BaseETS._resolve_q(), the single place that disambiguates the two shapes:
  • q of length ets.n is compact and gets scattered into a global-length array via ets.jindices
  • q of length >= max(jindices) + 1 is global and passes through unchanged (this also preserves two pre-existing behaviours: accepting a q longer than strictly needed, and never reordering an already-global q even when jindices aren't in increasing order, e.g. after .inv())
  • anything else raises ValueError naming both accepted lengths
• Wired into eval/jacob0/jacobe/hessian0/hessiane (3D ETS) and eval/jacob0/jacobe (2D ETS2, extended for consistency).
• No C++ changes needed at all -- the resolution happens once, in Python, before the facade decides between the C++ extension and the pure-Python fallback; both keep receiving exactly the global-length q they always have. Verified numerically identical between the two paths (~1e-16) on a real branched robot (YuMi).
• Updated RobotKinematics.fkine()'s and ETS.eval()'s docstrings to explain the new dual-mode q explicitly, with a worked YuMi example.

Item 4 (final item) of the IK-solver cleanup plan (see claude-notes/ik-solver-cpp-python-divergence.md). Items 1-3 are PRs #649, #650, #652 (all merged).

Test plan

• New regression tests: compact-vs-global equivalence on a synthetic branched sub-chain (eval/jacob0/jacobe/hessian0/hessiane), too-short q raises ValueError, trajectory support unaffected
• New end-to-end test on YuMi: ikine_LM -> fkine/jacob0 with the raw 7-element solution, no manual full-vector workaround
• Full test suite green (750 passed, 18 skipped) in a clean isolated venv, before and after merging in main's subsequent PRs
• Manually re-ran the 379.py repro end-to-end -- fkine(sol.q, end=...) now matches the target directly
• Sphinx docs build (-W --keep-going) clean of any new warnings, including the new runblock example in fkine's docstring
• Single-chain robots (Panda, Puma560, UR5) and the ETS.inv() reversed-jindex edge case confirmed unaffected

🤖 Generated with Claude Code

[codecov]

Codecov Report

❌ Patch coverage is 0% with 25 lines in your changes missing coverage. Please review.
✅ Project coverage is 0.00%. Comparing base (07a60f4) to head (b520774).

Files with missing lines	Patch %	Lines
src/roboticstoolbox/ets/_ETS.py	0.00%	15 Missing ⚠️
src/roboticstoolbox/ets/ETS.py	0.00%	7 Missing ⚠️
src/roboticstoolbox/ets/ETS2.py	0.00%	3 Missing ⚠️

Additional details and impacted files

@@          Coverage Diff          @@
##            main    #653   +/-   ##
=====================================
  Coverage   0.00%   0.00%           
=====================================
  Files        143     143           
  Lines      14027   14050   +23     
=====================================
- Misses     14027   14050   +23     

☔ 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.