feat(metrics): roll plugin_metrics up to hourly to bound storage

plugin_metrics grows by (sources × resources × ~30s cadence); keeping 90d of raw
is a large table. Add a raw→hourly rollup (mirroring the Docker plugin) so only a
short raw window is kept at full resolution, with hourly averages archived longer.

- PluginMetricHourly model + core migration 0024 (plugin_metrics_hourly: avg/max/
  count per source/resource/metric/hour, unique bucket constraint + lookup index).
- steward/core/metrics_retention.rollup_plugin_metrics: date_trunc('hour') agg of
  raw older than the hour-aligned raw window, idempotent pg upsert into hourly,
  delete the rolled raw, prune hourly beyond the rollup window.
- cleanup.py: plugin_metrics is no longer blanket-deleted at data.retention_days;
  _run_metrics_retention drives the rollup with windows read live from settings.
- Settings: metrics.retention.raw_days (7) + rollup_days (90), tunable on the
  Thresholds & Retention page (new "Host metrics retention" card).
- Chart read: _history_for_host merges the hourly rollup (older part of the range)
  with raw date_bin (recent part, capped ≤1h), so 30d charts keep working —
  recent at full resolution, older at hourly. Route passes raw_days from settings.
- Tests: unit (cutoff helpers) + integration (rollup aggregates/prunes; history
  merges hourly + raw) against Postgres.

Speed was already handled by the indexes + SQL aggregation; this is the storage
lever (raw window ~10x smaller).

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_016Jg27rgypiW2efULXJDtMC

steward/core/cleanup.py

@@ -6,7 +6,6 @@ from typing import TYPE_CHECKING
 from sqlalchemy import delete
 
 from steward.models.monitors import MonitorResult
-from steward.models.metrics import PluginMetric
 from steward.models.ansible import AnsibleRun
 
 if TYPE_CHECKING:
@@ -26,7 +25,6 @@ async def run_cleanup(app: "Quart") -> None:
         async with session.begin():
             for model, ts_col in [
                 (MonitorResult, MonitorResult.checked_at),
-                (PluginMetric, PluginMetric.recorded_at),
                 (AnsibleRun, AnsibleRun.started_at),
             ]:
                 result = await session.execute(
@@ -35,9 +33,29 @@ async def run_cleanup(app: "Quart") -> None:
                 if result.rowcount:
                     logger.info(f"Pruned {result.rowcount} rows from {model.__tablename__}")
 
+        # plugin_metrics is NOT blanket-deleted here — it's rolled up to hourly
+        # then pruned, so multi-week host history stays cheap.
+        await _run_metrics_retention(session, now)
         await _run_docker_retention(session, now)
 
 
+async def _run_metrics_retention(session, now: datetime) -> None:
+    """Roll up + prune plugin_metrics (raw → hourly → gone). Windows read fresh
+    from settings each run (rule 25 — UI change takes effect next cleanup, no
+    restart). get_setting's SELECT autobegins, so read inside the begin block."""
+    from steward.core.metrics_retention import rollup_plugin_metrics
+    from steward.core.settings import get_setting
+
+    async with session.begin():
+        raw_days = int(await get_setting(session, "metrics.retention.raw_days") or 7)
+        rollup_days = int(await get_setting(session, "metrics.retention.rollup_days") or 90)
+        counts = await rollup_plugin_metrics(
+            session, raw_days=raw_days, rollup_days=rollup_days, now=now,
+        )
+    if counts and any(counts.values()):
+        logger.info("Metrics retention: %s", counts)
+
+
 async def _run_docker_retention(session, now: datetime) -> None:
     """Drive the docker plugin's rollup + prune via its capability, if loaded.