locomotion.md: the gimp audio verdict (servo loop, no voice line) + rebind watch item

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

context/locomotion.md

@@ -303,6 +303,17 @@ gimped mechs too — the port relocated it into the NORMAL leg driver's Standing
 stops running when the gimp driver takes over ("rotating statue"). The arming is now mirrored
 in `AdvanceLegAnimationGimp`. No gimp-turn clip exists — a limping mech step-turns with the
 normal `trn` clip [T1: the RES limp set is only wg/gg/gs].
+**Gimp audio — RESOLVED 2026-07-29** (`ed84445`): there is NO "reverse disabled" voice line,
+klaxon binding, or MFD text in the 1995 game — the binary's mapper has no gimp check (the
+refusal is structural: the gimp machines have no reverse path) and no REVERSE/DISABLED string
+exists in BTL4.RES; the five `*VOC` bank samples are mission-shell VO. The AUTHENTIC limp audio
+is the authored **strained-servo component**: `Entity.AnimationState` state-watchers on the limp
+entries 22/23 and stops 26/27 (+ walk-stands 8/9) start it, gain-scaled by `Entity.LocalVelocity`
+(0..10 → −300..0). Verified live: `[statefire] trigState=23 old=7 new=23 ctl Start`
+(BT_AUDIO_SPATIAL diagnostic). ⚠ Open hazard: one bench session had ZERO AnimationState watcher
+fires (same build/conditions) — suspected watcher rebind miss around the initial mission respawn
+(watchers possibly bound to a recreated mech's stale indicator). Watch for silent limps in the
+field.
 Open: the load-time per-clip callbacks `PTR_LAB_0050d738/744` on the wg clips (uncarved LABs,
 inert `Recon` placeholders in the port — likely footstep/cycle events, transitions work without
-them); the gimp audio cue binding.
+them).