clean: archive legacy stt/tts/llm services; update ARCHITECTURE.md + README.md to current stack (REST gpt-4o-transcribe, nano, sage, Honcho, incremental TTS, white noise)

Per PLAN item 6. Legacy files moved to archive/legacy-pipeline/.

archive/legacy-pipeline/llm.py

@@ -0,0 +1,30 @@
+"""LLM service — DeepSeek API"""
+
+import logging
+from openai import AsyncOpenAI
+from config import settings
+
+logger = logging.getLogger("kira.llm")
+
+
+def _get_client() -> AsyncOpenAI:
+    return AsyncOpenAI(
+        api_key=settings.deepseek_api_key,
+        base_url=settings.deepseek_base_url,
+    )
+
+
+async def get_kira_response(messages: list[dict]) -> str:
+    """Get Kira's response from the LLM."""
+    try:
+        client = _get_client()
+        resp = await client.chat.completions.create(
+            model=settings.deepseek_model,
+            messages=messages,
+            max_tokens=300,
+            temperature=0.7,
+        )
+        return resp.choices[0].message.content or "Mhm, I'm here!"
+    except Exception as e:
+        logger.error(f"LLM error: {e}")
+        return "I'm still here with you! Could you say that again?"