feat(production): sync all modified production files to git

Includes updates across gateway, router, node-worker, memory-service,
aurora-service, swapper, sofiia-console UI and node2 infrastructure:

- gateway-bot: Dockerfile, http_api.py, druid/aistalk prompts, doc_service
- services/router: main.py, router-config.yml, fabric_metrics, memory_retrieval,
  offload_client, prompt_builder
- services/node-worker: worker.py, main.py, config.py, fabric_metrics
- services/memory-service: Dockerfile, database.py, main.py, requirements
- services/aurora-service: main.py (+399), kling.py, quality_report.py
- services/swapper-service: main.py, swapper_config_node2.yaml
- services/sofiia-console: static/index.html (console UI update)
- config: agent_registry, crewai_agents/teams, router_agents
- ops/fabric_preflight.sh: updated preflight checks
- router-config.yml, docker-compose.node2.yml: infra updates
- docs: NODA1-AGENT-ARCHITECTURE, fabric_contract updated

Made-with: Cursor

services/node-worker/main.py

@@ -43,7 +43,30 @@ async def prom_metrics():
 
 @app.get("/caps")
 async def caps():
-    """Capability flags for NCS to aggregate."""
+    """Capability flags for NCS to aggregate.
+
+    Semantic vs operational separation (contract):
+    - capabilities.voice_* = semantic availability (provider configured).
+      True as long as the provider is configured, regardless of NATS state.
+      Routing decisions are based on this.
+    - runtime.nats_subscriptions.voice_* = operational (NATS sub active).
+      Used for health/telemetry only — NOT for routing.
+
+    This prevents false-negatives during reconnects / restart races.
+    """
+    import worker as _w
+    nid = config.NODE_ID.lower()
+
+    # Semantic: provider configured → capability is available
+    voice_tts_cap = config.TTS_PROVIDER != "none"
+    voice_stt_cap = config.STT_PROVIDER != "none"
+    voice_llm_cap = True  # LLM always available when node-worker is up
+
+    # Operational: actual NATS subscription state (health/telemetry only)
+    nats_voice_tts_active = f"node.{nid}.voice.tts.request" in _w._VOICE_SUBJECTS
+    nats_voice_stt_active = f"node.{nid}.voice.stt.request" in _w._VOICE_SUBJECTS
+    nats_voice_llm_active = f"node.{nid}.voice.llm.request" in _w._VOICE_SUBJECTS
+
     return {
         "node_id": config.NODE_ID,
         "capabilities": {
@@ -53,6 +76,10 @@ async def caps():
             "tts": config.TTS_PROVIDER != "none",
             "ocr": config.OCR_PROVIDER != "none",
             "image": config.IMAGE_PROVIDER != "none",
+            # Voice HA semantic capability flags (provider-based, not NATS-based)
+            "voice_tts": voice_tts_cap,
+            "voice_llm": voice_llm_cap,
+            "voice_stt": voice_stt_cap,
         },
         "providers": {
             "stt": config.STT_PROVIDER,
@@ -65,6 +92,19 @@ async def caps():
             "vision": config.DEFAULT_VISION,
         },
         "concurrency": config.MAX_CONCURRENCY,
+        "voice_concurrency": {
+            "voice_tts": config.VOICE_MAX_CONCURRENT_TTS,
+            "voice_llm": config.VOICE_MAX_CONCURRENT_LLM,
+            "voice_stt": config.VOICE_MAX_CONCURRENT_STT,
+        },
+        # Operational NATS subscription state — for health/monitoring only
+        "runtime": {
+            "nats_subscriptions": {
+                "voice_tts_active": nats_voice_tts_active,
+                "voice_stt_active": nats_voice_stt_active,
+                "voice_llm_active": nats_voice_llm_active,
+            }
+        },
     }