v0.13.0:4 - redaction gateway, embeddings proxy, expanded audio API

- Add redaction gateway (redaction_gateway.py, redaction/ scrub + tests)
- Add embeddings proxy and spark_embed service (Dockerfile + main.py)
- Expand audio_proxy with speaker-aware handling; deep_health/health/server updates
- Package: configureSparks action + sparkConfig model updates, manifest/main wiring
- Docs: AUDIO_API, EMBEDDINGS, REDACTION_GATEWAY; HANDOFF and runbook/known-issues refresh

image/app/llm_proxy.py

@@ -1,10 +1,10 @@
 """OpenAI-compatible chat-completions proxy that forwards to the vLLM
 process currently running on Spark 1.
 
-Lets clients (recap-relay, Open WebUI, etc.) use a single Spark Control
+Lets clients (Open WebUI, custom apps, etc.) use a single Spark Control
 host for everything — same TLS cert, same allowlist, same place to add
 rate limiting/observability later — instead of having to also reach
-into <spark-1-ip>:8888 directly.
+into <spark1-host>:8888 directly.
 
 Endpoints:
   POST /v1/chat/completions   — OpenAI chat completions (streams when stream=true)