docs: simplify integration path section

Author: Nehanth

docs/claude-agent/mlflow-tracing.md

@@ -47,11 +47,7 @@ This works the same whether the backend is Vertex AI, vLLM directly, or OGX →
 
 ### Integration Path
 
-| Path | Status |
-|---|---|
-| **Agent-level → MLflow** (`mlflow autolog claude`) | ✅ Working — tested across all three backends |
-| OGX → custom exporter → MLflow | 🔲 Future — would add server-side signals on top |
-| OGX → OTel → MLflow | 🔲 Future — requires OTel support in both OGX and MLflow |
+The Claude Code stop hook is the right integration path. It already captures everything out of the box — tool calls, token usage, latency, session ID — and works the same across Vertex AI, vLLM, and OGX without any changes. If additional server-side metrics are needed (e.g. per-hop vLLM latency, OGX routing decisions), they can be added directly to the same hook since the infrastructure is already there.
 
 ### Evidence: Same Traces Across All Three Backends