chore: promote develop to master (yolo mps ultralytics fallback fix)

[solderzzc]

Summary

Promotes develop to master. The relevant change for this PR is #210 (already merged into develop), which needs to reach master because the Aegis-AI app's skill installer clones the yolo-detection-2026 skill from master at install/reinstall time (skill-registry-service.cjs) — develop isn't in that path, so #210 alone didn't reach any installed skill.

• load_optimized()'s final fallback branch (taken when the optimized runtime check fails, framework_ok=False) unconditionally imported ultralytics — but mps installs deliberately never ship torch/ultralytics (requirements_mps.txt only lists onnxruntime).
• Any mps machine where _check_mps_runtime() fails (CoreML execution provider not found) crashed with No module named 'ultralytics' when loading a YOLO skill, instead of falling back to the pre-built .onnx model via plain CPUExecutionProvider.
• Reported in SharpAI/DeepCamera#207.

Locally dry-ran this merge (git merge origin/develop onto master, no --no-ff conflicts) — clean, no conflicts; the diff is exactly the #210 fix (104 lines across 2 env_config.py copies + 1 new test file), nothing else from develop leaks in unexpectedly since develop's only commits ahead of master are the #210 fix itself.

Test plan

• skills/lib/test_env_config_mps_fallback.py (3 tests) re-verified against current master post-merge — passes.
• Full skills/lib/test_env_config_*.py suite — 4 pre-existing failures in test_env_config_ane.py reproduce identically on master without this change (confirmed by temporarily reverting just the two env_config.py files to master's version and re-running) — not introduced by this change.
• Not yet verified against Brian's actual machine — this PR unblocks him reinstalling the yolo-detection-2026 skill from the Aegis-AI Skills sidebar to pick it up.

🤖 Generated with Claude Code