Move _head_proto into blender_assets.py and add assemble --head-proto; verify prototype weights

TODO.md

@@ -97,8 +97,15 @@ the rest of §C remains the broad review checklist.
       and 1.5 m, front/three-quarter/profile. Vision-judged: **s48 wins** — s24 is too faceted
       (nose/lips melt, jagged silhouette), s64 is visually identical to s48 at both distances
       (extra ~900 verts wasted). Rejected variants stay in `tmp/npc-parts/`.
-- [ ] Verify head weights stay on the Head bone, headgear clearance still passes `fit`, and the
+- [-] Verify head weights stay on the Head bone, headgear clearance still passes `fit`, and the
       face does not bake recessed creases into dark AO patches.
+      Weight check done: `npc_parts.py assemble --head-proto` passes `inspect` (npc_base, 14 bones,
+      skinned, no issues) and the head join is weighted 100% Head by construction
+      (`rigid=lambda p: "Head"`). Headgear clearance and the AO no-blotch check are pending the
+      §5 rollout (the `fit`/`bake_ao` tools read the committed assets; the ad-hoc AO bake on the
+      imported GLB came out black and is superseded by the proper `.blend`-source bake). Flag: the
+      prototype ears reach x ±0.162 vs the 0.1575 head-radius reference, so headgear may clip and
+      must be re-measured at rollout.
 - [ ] Get owner sign-off on the one-variant prototype, then roll the approved head/face into the
       other 11 variants one asset family at a time, with a before/after sheet per batch.