Add ask_expert and eval capture (A14/A16) — step 7c

Completes step 7.

- expert.py: one stateless call for a question needing no repo access.
  Structured output is contracted at the prompt level, since vendors
  differ on JSON enforcement, so the reply is parsed and validated
  locally with one retry — tolerating a fenced code block, because
  models wrap JSON habitually, but not missing fields
- The per-task budget is counted from the canonical log rather than
  memory, so it survives a restart and cannot hand a circling executor
  a fresh allowance. Exhaustion raises rather than answering: a caller
  out of consults should get on with the work or escalate
- Consults are recorded as delegation_started/completed with role
  consultant. A16 asks for a "consult/delegation eval set", and
  reusing the composite events gets consult evaluation for free rather
  than inventing a parallel record
- evals.py pairs starts with completions into eval records, with
  verdicts kept outside the canonical log: the log records what
  happened, a verdict is a later judgement about it, and judgements
  get revised. Schema v3 adds the outcomes table, so a v1 store now
  traverses two migrations — a better exercise of the hook than one
- ask_expert is offered to the executor as a tool only when a desk is
  configured, in addition to the observable escalation triggers and
  never instead of them (A12). Consulting is not mutating, so a
  completion claim backed only by consults is still challenged

Found live: passing an empty message list and putting the question in
the system prompt fails outright — the Messages API requires at least
one non-system message. The question now travels as a synthesised user
message that is never appended to the log, which keeps the consult
stateless while remaining a valid request.

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

tests/integration/test_expert_live.py

@@ -0,0 +1,60 @@
+"""A real consult over the normalized route (A14).
+
+The point is the prompt-level contract: a live model must produce something
+that parses against our local schema, since no API-level enforcement exists.
+"""
+
+import io
+import os
+from pathlib import Path
+
+import pytest
+
+from nisse_dev.envfile import load_env_file
+from nisse_dev.evals import collect
+from nisse_dev.expert import BudgetExhausted, ExpertDesk
+from nisse_dev.llm import LiteLLMClient, RoleMap
+from nisse_dev.oplog import OpLog
+from nisse_dev.state import StateStore
+
+load_env_file(Path(__file__).resolve().parents[2] / ".env")
+
+SENTINEL = "NISSE_LIVE_PLANNER_MODEL"
+
+pytestmark = [
+    pytest.mark.integration,
+    pytest.mark.skipif(
+        not os.environ.get(SENTINEL),
+        reason=f"no live frontier route configured ({SENTINEL} unset)",
+    ),
+]
+
+
+def desk(tmp_path: Path, budget: int = 2) -> tuple[ExpertDesk, StateStore]:
+    model = os.environ[SENTINEL]
+    roles = RoleMap(executor=model, planner=model, delegate=model, consultant=model)
+    store = StateStore(tmp_path / "state.db", oplog=OpLog("nisse-dev", io.StringIO()))
+    return ExpertDesk(LiteLLMClient.from_env(), store, roles, budget=budget), store
+
+
+def test_live_consult_parses_against_the_local_schema(tmp_path: Path) -> None:
+    expert, store = desk(tmp_path)
+
+    answer = expert.ask(
+        "For a single-process agent storing a few thousand rows of task state, "
+        "is SQLite sufficient, or is PostgreSQL warranted? Answer briefly."
+    )
+
+    assert answer.confidence in {"high", "medium", "low"}
+    assert len(answer.answer) > 20
+    # The consult is captured for A16 without any extra bookkeeping.
+    [record] = collect(store)
+    assert record.role == "consultant"
+    assert record.exit_status == "answered"
+
+
+def test_live_budget_stops_a_circling_caller(tmp_path: Path) -> None:
+    expert, _ = desk(tmp_path, budget=1)
+    expert.ask("Name one advantage of WAL mode in SQLite. Be brief.")
+    with pytest.raises(BudgetExhausted):
+        expert.ask("And another?")