Add tts/text output schema

engine/docs/ws_v1_schema.md

@@ -53,6 +53,9 @@ Rules:
   },
   "metadata": {
     "client": "web-debug",
+    "output": {
+      "mode": "audio"
+    },
     "systemPrompt": "You are concise.",
     "greeting": "Hi, how can I help?",
     "services": {
@@ -70,6 +73,7 @@ Rules:
         "minAudioMs": 300
       },
       "tts": {
+        "enabled": true,
         "provider": "siliconflow",
         "model": "FunAudioLLM/CosyVoice2-0.5B",
         "apiKey": "sf-...",
@@ -83,6 +87,10 @@ Rules:
 
 `metadata.services` is optional. If omitted, server defaults to environment configuration.
 
+Text-only mode:
+- Set `metadata.output.mode = "text"` OR `metadata.services.tts.enabled = false`.
+- In this mode server still sends `assistant.response.delta/final`, but will not emit audio frames or `output.audio.start/end`.
+
 ### `input.text`
 
 ```json