Superuser route to flush the metrics collector now

POST /observability/flush handles what the collector has queued and writes
it, so the soak harness no longer sleeps FLUSH_INTERVAL_S before clearing
its own rows.

backend/fluksio/api/routes/observability.py

@@ -19,12 +19,17 @@ from sqlalchemy import select as sa_select
 from sqlmodel import col, select
 
 from fluksio import __version__
-from fluksio.api.deps import FlowControllerDep, SessionDep, get_current_user
+from fluksio.api.deps import (
+    FlowControllerDep,
+    SessionDep,
+    get_current_active_superuser,
+    get_current_user,
+)
 from fluksio.api.routes.runs import elapsed_ms
 from fluksio.core.config import settings
 from fluksio.flow.controller import NodeStatus
 from fluksio.flow.pipeline import ADVISORY_ISSUES
-from fluksio.models import EngineEvent, FlowRun, MetricBucket
+from fluksio.models import EngineEvent, FlowRun, Message, MetricBucket
 
 router = APIRouter(
     prefix="/observability",
@@ -508,3 +513,18 @@ async def read_dead_letters(
     if controller.execution is None:
         return []
     return await run_in_threadpool(controller.execution.queue.dead_letters, limit)
+
+
+@router.post(
+    "/flush",
+    response_model=Message,
+    dependencies=[Depends(get_current_active_superuser)],
+)
+async def flush_metrics(request: Request) -> Any:
+    """Write what the metrics collector holds now, not at its next interval.
+
+    For tooling that reads or clears these tables right after causing events,
+    such as the soak harness clearing its own rows.
+    """
+    await request.app.state.metrics_collector.flush_now()
+    return Message(message="Metrics flushed")