feat(M5-M7): embed canonical material names in USD via customData + pxr direct read

- export_step_to_usd.py: accept --material_map CLI arg, write
  schaeffler:canonicalMaterialName as customData on each Mesh prim,
  fix geometry transform (strip shape Location before face exploration,
  apply both face_loc and shape_loc sequentially)
- import_usd.py: after Blender USD import, use pxr to read customData
  directly from the USD file — builds {part_key: material_name} lookup
  (Blender ignores STRING primvars and customData, but pxr reads both)
- _blender_materials.py: add apply_material_library_direct() for exact
  dict-based material assignment without name-matching heuristics
- _blender_scene_setup.py: prefer direct USD lookup, fall back to
  name-matching for legacy USD files without material metadata
- export_glb.py (generate_usd_master_task): resolve material_map via
  material_service.resolve_material_map() and pass to subprocess;
  include material hash in cache key for invalidation
- ROADMAP.md: update P5 status, add M5-M7 milestones

Tested: 3/3 parts matched (ans_lfs120), 172/175 parts matched
(F-802007.TR4-D1-H122AG). Previous: 0/25 matched.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

.claude/commands/review.md

@@ -25,7 +25,7 @@ You are the reviewer for the Schaeffler Automat project. You check implemented c
 - [ ] `storage_key` values are relative (never start with `/`)
 
 ### Celery / Tasks
-- [ ] Task is on the correct queue? (`thumbnail_rendering` for ALL Blender/render-worker calls)
+- [ ] Task is on the correct queue? (`asset_pipeline` for ALL Blender/render-worker calls)
 - [ ] No Blender/subprocess call on `step_processing` queue
 - [ ] `self.request.id` written to `render_job_doc.celery_task_id` at task start
 - [ ] `PipelineLogger` used for step start/done/error events