Stage 6: learn_call_flow rebuilt on the learner, docs truth sweep

The call-flow learner finally gets fed: exploration mode records its
IVR discoveries on the call (ActiveCall.exploration_steps) instead of
throwing them away, persistence stores them in the call record's
metadata, and the rebuilt learn_call_flow MCP tool turns a completed
exploration call into a stored flow via CallFlowLearner — correct
constructor (llm_client from get_llm, heuristic labels when the LLM
is unavailable), build for a new number, merge/refine when a flow
already exists. save_learned_flow/update_flow_from_model keep the
CallFlow↔row mapping in call_persistence.

Test gaps closed: tests/test_learner.py (discoveries→linked steps,
exploration persistence, learn-then-refine through the in-memory MCP
client, no-data and unknown-call answers) and tests/test_websocket.py
(4401 without token, trunk-status-then-replay on connect, per-call
stream filtering).

Docs aligned to code: README (15 tools incl. learn_call_flow, HTTP
not SSE, Python 3.12+, PostgreSQL+Alembic — no SQLite fallback, media
pipeline marked stub-mode until pjsua2 installed, Alembic and honest
/health checked off); docs/mcp-server.md rewritten against the actual
tool surface (hangup not end_call, real params, 3 real resources,
/mcp/ streamable HTTP + bearer auth); architecture/development/
configuration drift fixed.

pyproject: pruned never-imported deps (websockets, librosa,
soundfile, python-multipart).

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

services/call_persistence.py

@@ -136,6 +136,36 @@ async def create_flow(
     return row
 
 
+async def save_learned_flow(session: AsyncSession, flow: CallFlow) -> StoredCallFlow:
+    """The one CallFlow-model → row mapping (auto-learned flows)."""
+    row = StoredCallFlow(
+        id=flow.id,
+        name=flow.name,
+        phone_number=flow.phone_number,
+        description=flow.description,
+        steps=[s.model_dump(mode="json") for s in flow.steps],
+        tags=flow.tags,
+        notes=flow.notes,
+        times_used=flow.times_used,
+        last_used=flow.last_used,
+        last_verified=datetime.now(),
+    )
+    session.add(row)
+    await session.flush()
+    return row
+
+
+async def update_flow_from_model(
+    session: AsyncSession, row: StoredCallFlow, flow: CallFlow
+) -> None:
+    """Write a refined CallFlow back onto its existing row."""
+    row.steps = [s.model_dump(mode="json") for s in flow.steps]
+    row.times_used = flow.times_used
+    row.last_used = flow.last_used
+    row.notes = flow.notes
+    await session.flush()
+
+
 # ================================================================
 # Devices
 # ================================================================
@@ -320,7 +350,10 @@ async def _finalize_call_record(call: ActiveCall, final_status: CallStatus) -> N
             }
             for c in call.classification_history
         ]
-        record.metadata_ = {"services": list(call.services)}
+        metadata = {"services": list(call.services)}
+        if call.exploration_steps:
+            metadata["exploration_steps"] = call.exploration_steps
+        record.metadata_ = metadata
 
         # Each transcript entry gets its own row with a sequence number
         # and real offset so the dashboard can render click-to-seek.

services/hold_slayer.py

@@ -294,7 +294,7 @@ class HoldSlayerService:
         logger.info(f"🔍 Exploration mode: discovering IVR for {call.remote_number}")
         await self.call_manager.update_status(call.id, CallStatus.NAVIGATING_IVR)
 
-        discovered_steps: list[dict] = []
+        discovered_steps = call.exploration_steps  # persisted with the record
         max_time = self.settings.hold_slayer.max_hold_time
         start_time = time.time()