IB-WP-0019-T02: usage rollup from run records

Every completed generate run now aggregates per-call adapter usage from
the workflow-engine run records into output/budget/usage.yaml. Per-call
data is bucketed by (workflow_id, stage_id, provider, model) with
running totals for calls, prompt_tokens, completion_tokens,
total_tokens, and cost_usd_known (sum of adapter-reported cost when the
provider returns it; usually zero today). A run-level entry captures
run_index, started_at, completed_at, duration_seconds, the executing
plan snapshot_id (resolved from the latest plans.yaml entry), and the
workflow-level run_id / stage_count summaries.

cost_usd_estimated is left as None for this task; T03 wires the
rate-table resolver so the same bucket gets a model-priced fallback
when the adapter does not return cost directly.

Fixture-mode runs are recorded with provider='fixture', zero tokens,
and cost_status='unknown' rather than silently skipped, so the rollup
honestly reflects which stages actually ran.

102 tests pass.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

src/infospace_bench/generator.py

@@ -2,11 +2,14 @@ from __future__ import annotations
 
 import hashlib
 import shutil
+import time
 from dataclasses import asdict, dataclass, field
 from datetime import datetime, timezone
 from pathlib import Path
 from typing import Any
 
+_monotonic = time.monotonic
+
 import yaml
 
 from .checks import run_collection_checks
@@ -15,7 +18,11 @@ from .evaluation_io import read_entity_evaluations
 from .history import get_history, read_metrics_file, record_check_results
 from .lifecycle import create_infospace, load_infospace, register_artifact
 from .openrouter import OpenRouterAssistedGenerationAdapter
-from .budget import record_plan_snapshot
+from .budget import (
+    latest_plan_snapshot_id,
+    record_plan_snapshot,
+    record_run_usage,
+)
 from .source_intake import SourceChunk, normalize_source
 from .workflow import (
     AssistedGenerationAdapter,
@@ -343,6 +350,8 @@ def run_generation(
             metrics=status.get("metrics", {}),
         )
 
+    started_wall = datetime.now(timezone.utc)
+    monotonic_start = _monotonic()
     adapter = (
         _adapter_for(provider, model=model, fixture_responses=fixture_responses)
         if workflow_ids
@@ -379,6 +388,15 @@ def run_generation(
         }
     )
     _write_state(root_path, state)
+    if workflow_results:
+        duration_seconds = round(_monotonic() - monotonic_start, 3)
+        record_run_usage(
+            root_path,
+            workflow_results,
+            snapshot_id=latest_plan_snapshot_id(root_path),
+            duration_seconds=duration_seconds,
+            started_at=started_wall.isoformat(),
+        )
     return GenerationRunResult(
         root=str(root_path),
         status="completed",