refactor(docker): container CPU and memory live in the one metric store (#5752)

#5752 "Container CPU and memory history lives in the one metric store;
docker_metrics and docker_metrics_hourly are retired". Container cpu_pct and
mem_pct already went to plugin_metrics for alerting; now mem_usage_bytes does
too, and every snapshot in an ingest (backlog included) is recorded at its
own time. The container history chart, Busiest Containers, a service's
history and the MCP container tools read plugin_metrics under resource
"<host>/<container>", rolled up and pruned by the shared metric retention.

docker_012_drop_metrics drops both tables. Their history is not copied
(lesson #5221 "A data migration's cost is set by the live table"); the
store already holds 7 days of container CPU and memory %. Docker retention
now prunes only lifecycle events, and Settings -> Thresholds has one
"Metric retention" card for every source.

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

tests/integration/test_docker_recent_usage.py

@@ -1,6 +1,7 @@
 """Integration: Busiest Containers bars average each container's samples per
-time bucket in one query (#5214). Two samples share a bucket and must merge
-(lesson #5262). Requires STEWARD_DATABASE_URL."""
+time bucket in one query (#5214), read from the one metric store (#5752). Two
+samples share a bucket and must merge (lesson #5262). Requires
+STEWARD_DATABASE_URL."""
 from __future__ import annotations
 
 import asyncio
@@ -33,7 +34,7 @@ def _docker(sub):
 
 def test_recent_usage_averages_each_containers_bucket(app):
     from steward.models.hosts import Host
-    DockerMetric = _docker("models").DockerMetric
+    from steward.models.metrics import PluginMetric
     service = _docker("service")
     bar_slots, recent_usage = service.bar_slots, service.recent_usage
 
@@ -43,16 +44,18 @@ def test_recent_usage_averages_each_containers_bucket(app):
     at = slots[-2]                                   # a whole bucket, in the past
 
     def m(name, secs, cpu, mem):
-        return DockerMetric(host_id=hid, container_name=name, scraped_at=at + timedelta(seconds=secs),
-                            cpu_pct=cpu, mem_pct=0.0, mem_usage_bytes=mem)
+        when = at + timedelta(seconds=secs)
+        return [PluginMetric(source_module="docker", resource_name=f"dock-{tag}/{name}",
+                             metric_name=metric, value=value, recorded_at=when)
+                for metric, value in (("cpu_pct", cpu), ("mem_usage_bytes", mem))]
 
     async def go():
         async with app.db_sessionmaker() as s:
             async with s.begin():
                 s.add(Host(id=hid, name=f"dock-{tag}", address="10.9.9.4"))
             async with s.begin():
-                s.add_all([m(f"a-{tag}", 10, 100.0, 1000), m(f"a-{tag}", 70, 300.0, 3000),
-                           m(f"b-{tag}", 10, 50.0, 500)])
+                s.add_all(m(f"a-{tag}", 10, 100.0, 1000) + m(f"a-{tag}", 70, 300.0, 3000)
+                          + m(f"b-{tag}", 10, 50.0, 500))
             async with s.begin():
                 return await recent_usage(s, {(hid, f"a-{tag}")}, slots[0], width)