feat(norms): Steward learns each metric's normal per hour of day, then week (#5739)

A metric registered with learn_normal() gets, for every resource reporting
it, a usual hold (the 90th percentile across past days of each hour's p95)
and a usual peak for each slot:
- an hour of the day once 5 days of history exist;
- an hour of the week once history reaches back 3 weeks, so weekly jobs are
  learned too.
Each slot takes in the hours either side, so a job that runs a little late
stays normal. One SQL statement relearns everything hourly into
metric_norms, from the rollup plus the raw window not rolled up yet;
norms_at() reads the slot a moment falls in. Nothing is registered yet:
Traefik (#5741) and hosts (#5742) opt in.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

steward/app.py

@@ -445,6 +445,10 @@ def _register_core_tasks(app: Quart) -> None:
         from .backup.job import tick
         await tick(app)
 
+    async def run_metric_norms():
+        from .core.norms import run_norms
+        await run_norms(app)
+
     async def run_email_outbox():
         from .core.outbox import send_due
         await send_due(app)
@@ -507,6 +511,15 @@ def _register_core_tasks(app: Quart) -> None:
             interval_seconds=cleanup_interval,
             run_on_startup=False,
         ),
+        # Each learned metric's normal per hour (#5739). Hourly, since a slot
+        # gains one hour of history an hour; the table outlives a restart, so
+        # there is nothing to rebuild at startup.
+        ScheduledTask(
+            name="metric_norms",
+            coro_factory=run_metric_norms,
+            interval_seconds=3600,
+            run_on_startup=False,
+        ),
     ])
 
     # Pull Ansible git sources as each falls due. One task reads the sources