feat: no-answer retry, agent status machine, escalation events/timeline (ТЗ §13-15,22,25-27,37, AC-08,16-19)

Phase 2 of the L1->L2 routing engine (Phase 1: MR!4).

- ami_loop() now also captures native AMI DialEnd/Hangup frames (not
  only UserEvent), needed to detect that an escalated agent did not
  answer. No dialplan change required - Redirect already routes the
  client channel into an existing Dial()-based transfer context, so
  Asterisk emits these events on its own; the listener just wasn't
  reading them before.
- retry_escalation_no_answer(): on NOANSWER/BUSY/CANCEL/CHANUNAVAIL/
  CONGESTION, releases the non-answering agent, excludes it, and
  reserves+redirects to the next available agent via the routing
  engine's existing exclude_agent_ids support. Exhausted pool marks
  the escalation failed and leaves the call with the AI instead of
  dropping the client (ТЗ §32).
- Agent status now actually moves through
  RESERVED -> RINGING -> TALKING -> AFTER_CALL_WORK -> AVAILABLE
  instead of staying stuck on RESERVED for the whole call; a new
  acw_sweep_loop background thread (same pattern as the existing
  failed_retry_loop) times out AFTER_CALL_WORK back to AVAILABLE.
- escalations gains attempt_count/real_agent_id/attempted_agent_ids_json
  (migration 0033); fixes a latent bug where assigned_agent_id stored
  the SIP extension instead of the real agent_id despite routing-service
  already returning it in RoutingAgentReserveOut.
- Every transition now records an interaction timeline entry and
  publishes the ТЗ §25 event catalog (AgentReserved/AgentRinging/
  AgentNoAnswer/AgentConnected/TransferCompleted/TransferFailed)
  through the existing emit_voice_event/EventOutboxRow idempotent path.

Not in this MR (see plan): SLA config, Callback, L3 (needs real
technical agents from the business), metrics.

tests/test_escalation_retry.py

@@ -0,0 +1,190 @@
+import json
+
+import pytest
+
+import services.asterisk_bridge_service.voice_ai as voice_ai
+from services.routing_service import engine as routing_engine
+from services.shared.core import new_id, utc_now_iso
+from services.shared.db import get_session
+from services.shared.sql_init import init_sql_schema
+from services.shared.sql_models import AgentRow, AsteriskCallLinkRow, EscalationRow
+
+
+def _make_agent(session, *, extension: str, status: str = "AVAILABLE"):
+    now = utc_now_iso()
+    row = AgentRow(
+        agent_id=new_id("agt"),
+        tenant_ids_json="[]",
+        extension=extension,
+        endpoint=None,
+        display_name=f"Agent {extension}",
+        level="L2",
+        skills_json="[]",
+        status=status,
+        max_concurrent_calls=1,
+        enabled=True,
+        created_at=now,
+        updated_at=now,
+    )
+    session.add(row)
+    session.commit()
+    session.refresh(row)
+    return row
+
+
+def _make_call_link(session, *, call_id: str):
+    now = utc_now_iso()
+    link = AsteriskCallLinkRow(
+        call_id=call_id,
+        linked_id=call_id,
+        queue_code="voice_lab_ai",
+        queue_id="que_test",
+        interaction_id="int_test",
+        status="active",
+        telephony_status="ringing",
+        channel_name=f"PJSIP/tele2-kazgaz-{call_id}",
+        current_level="L2",
+        started_at=now,
+        updated_at=now,
+    )
+    session.add(link)
+    session.commit()
+    session.refresh(link)
+    return link
+
+
+def _make_ringing_escalation(session, *, call_id: str, agent: AgentRow):
+    now = utc_now_iso()
+    escalation = EscalationRow(
+        escalation_id=new_id("esc"),
+        call_id=call_id,
+        tenant_id=None,
+        from_level="L1",
+        to_level="L2",
+        reason_code="AI_UNABLE_TO_RESOLVE",
+        required_skills_json="[]",
+        priority=3,
+        status="ringing",
+        assigned_agent_id=agent.extension,
+        real_agent_id=agent.agent_id,
+        attempted_agent_ids_json=json.dumps([agent.agent_id]),
+        requested_at=now,
+    )
+    session.add(escalation)
+    session.commit()
+    session.refresh(escalation)
+    return escalation
+
+
+def _patch_routing_over_http(monkeypatch, session):
+    """Make voice_ai's HTTP-facing routing helpers operate on the same test session directly."""
+
+    def fake_reserve(*, call_id, level, tenant_id, required_skills=None, exclude_agent_ids=None):
+        agent = routing_engine.reserve_agent(
+            session,
+            call_id=call_id,
+            level=level,
+            tenant_id=tenant_id,
+            required_skills=required_skills,
+            exclude_agent_ids=exclude_agent_ids,
+        )
+        if agent is None:
+            return None
+        return {
+            "agent_id": agent.agent_id,
+            "extension": agent.extension,
+            "endpoint": agent.endpoint,
+            "display_name": agent.display_name,
+        }
+
+    def fake_release_by_agent_id(agent_id, *, next_status="AVAILABLE"):
+        routing_engine.release_agent_by_id(session, agent_id=agent_id, next_status=next_status)
+
+    def fake_set_status(agent_id, status):
+        routing_engine.set_agent_status(session, agent_id=agent_id, status=status)
+
+    monkeypatch.setattr(voice_ai, "_reserve_routing_agent", fake_reserve)
+    monkeypatch.setattr(voice_ai, "routing_release_by_agent_id", fake_release_by_agent_id)
+    monkeypatch.setattr(voice_ai, "set_routing_agent_status", fake_set_status)
+    monkeypatch.setattr(voice_ai, "_append_escalation_timeline", lambda *a, **k: None)
+    monkeypatch.setattr(voice_ai, "_emit_escalation_event", lambda *a, **k: None)
+    monkeypatch.setattr(voice_ai, "_resolve_handoff_channel", lambda session, link: link.channel_name)
+
+
+def test_retry_escalation_no_answer_moves_to_next_available_agent(monkeypatch):
+    init_sql_schema()
+    redirected_to: list[str] = []
+
+    session = get_session()
+    try:
+        _patch_routing_over_http(monkeypatch, session)
+        monkeypatch.setattr(
+            voice_ai,
+            "_redirect_channel_to_agent",
+            lambda *, channel, extension: redirected_to.append(extension),
+        )
+        agent_a = _make_agent(session, extension="2001")
+        agent_b = _make_agent(session, extension="2002")
+        routing_engine.reserve_agent(session, call_id="call-retry-1", level="L2", tenant_id=None, required_skills=[])
+        link = _make_call_link(session, call_id="call-retry-1")
+        escalation = _make_ringing_escalation(session, call_id="call-retry-1", agent=agent_a)
+
+        voice_ai.retry_escalation_no_answer(session, call_id="call-retry-1", dial_outcome="NOANSWER")
+
+        session.refresh(escalation)
+        session.refresh(agent_a)
+        session.refresh(agent_b)
+
+        assert escalation.attempt_count == 1
+        assert escalation.status == "ringing"
+        assert escalation.real_agent_id == agent_b.agent_id
+        assert escalation.assigned_agent_id == agent_b.extension
+        assert json.loads(escalation.attempted_agent_ids_json) == [agent_a.agent_id, agent_b.agent_id]
+
+        assert agent_a.status == "AVAILABLE"
+        assert agent_a.current_call_id is None
+        assert agent_b.status == "RINGING"
+        assert agent_b.current_call_id == "call-retry-1"
+
+        assert redirected_to == [agent_b.extension]
+    finally:
+        session.close()
+
+
+def test_retry_escalation_no_answer_exhausts_pool_marks_failed(monkeypatch):
+    init_sql_schema()
+
+    session = get_session()
+    try:
+        _patch_routing_over_http(monkeypatch, session)
+        monkeypatch.setattr(voice_ai, "_redirect_channel_to_agent", lambda **kwargs: None)
+        agent_a = _make_agent(session, extension="3001")
+        routing_engine.reserve_agent(session, call_id="call-retry-2", level="L2", tenant_id=None, required_skills=[])
+        link = _make_call_link(session, call_id="call-retry-2")
+        escalation = _make_ringing_escalation(session, call_id="call-retry-2", agent=agent_a)
+
+        voice_ai.retry_escalation_no_answer(session, call_id="call-retry-2", dial_outcome="NOANSWER")
+
+        session.refresh(escalation)
+        session.refresh(agent_a)
+
+        assert escalation.attempt_count == 1
+        assert escalation.status == "failed"
+        assert agent_a.status == "AVAILABLE"
+    finally:
+        session.close()
+
+
+def test_retry_escalation_no_answer_ignores_calls_without_ringing_escalation(monkeypatch):
+    init_sql_schema()
+    called = []
+
+    session = get_session()
+    try:
+        _patch_routing_over_http(monkeypatch, session)
+        monkeypatch.setattr(voice_ai, "_redirect_channel_to_agent", lambda **kwargs: called.append(kwargs))
+        _make_call_link(session, call_id="call-no-escalation")
+        voice_ai.retry_escalation_no_answer(session, call_id="call-no-escalation", dial_outcome="NOANSWER")
+        assert called == []
+    finally:
+        session.close()