Add binaural recording + tune spatial tracking

binaural_recorder.py: Records left/right ear streams as stereo WAV
in rolling 5-minute segments. Training data for spatial audio models.
Enabled via BINAURAL_RECORD=1 env var.

spatial.py: Tune smoothing — alpha 0.3→0.4 (snappier response),
idle return speed 0.05→0.03 (gentler drift), timeout 2s→1.5s.

headmic.py: Wire binaural recorder into audio loop, add /recording
endpoint for stats, feed both ear streams (not just best beam).

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

headmic.py

@@ -170,6 +170,9 @@ sound_ring_buffer = None  # collections.deque, filled by listener_loop
 # Speaker recognizer globals
 speaker_recognizer = None
 enrollment_buffer = None  # list of frame bytes, set during enrollment
+
+# Binaural recorder
+binaural_recorder = None
 enrollment_name = None
 
 # Audio stream
@@ -265,6 +268,13 @@ def listener_loop():
 
             state.active_side = side
 
+            # Feed binaural recorder (both ears)
+            if binaural_recorder:
+                binaural_recorder.feed(
+                    dual_stream.left.get_frame(),
+                    dual_stream.right.get_frame() if dual_stream.right else None
+                )
+
             # Feed sound classifier ring buffer
             if sound_ring_buffer is not None:
                 sound_ring_buffer.append(frame_data)
@@ -454,7 +464,7 @@ app = FastAPI(title="HeadMic", description="Vixy's Ears 🦊👂 (Dual XVF3800)"
 
 @app.on_event("startup")
 async def startup():
-    global sound_classifier, sound_ring_buffer, speaker_recognizer, dual_stream, LEDS_AVAILABLE, spatial_tracker
+    global sound_classifier, sound_ring_buffer, speaker_recognizer, dual_stream, LEDS_AVAILABLE, spatial_tracker, binaural_recorder
 
     state.running = True
 
@@ -530,6 +540,15 @@ async def startup():
         logger.info("Spatial tracking started (%d Hz, %.0fmm baseline, pushing gaze to %s)",
                      DOA_POLL_HZ, array_sep, EYE_SERVICE_URL)
 
+    # --- Binaural recording ---
+    if os.environ.get("BINAURAL_RECORD", "").lower() in ("1", "true", "yes"):
+        from binaural_recorder import BinauralRecorder
+        rec_dir = os.environ.get("BINAURAL_DIR", os.path.expanduser("~/headmic/recordings"))
+        binaural_recorder = BinauralRecorder(output_dir=rec_dir)
+        binaural_recorder.start()
+    else:
+        logger.info("Binaural recording disabled (set BINAURAL_RECORD=1 to enable)")
+
     # --- Main listener ---
     thread = threading.Thread(target=listener_loop, daemon=True)
     thread.start()
@@ -540,6 +559,8 @@ async def startup():
 async def shutdown():
     state.running = False
     leds_off()
+    if binaural_recorder:
+        binaural_recorder.stop()
     if dual_stream:
         dual_stream.stop()
 
@@ -606,6 +627,16 @@ async def doa():
     }
 
 
+# --- Binaural recording ---
+
+@app.get("/recording")
+async def recording():
+    """Binaural recording status."""
+    if not binaural_recorder:
+        return {"recording": False, "enabled": False}
+    return binaural_recorder.stats
+
+
 # --- Device info ---
 
 @app.get("/devices")