feat: DashboardDataGrid migration + FileList 7-feature rewrite + validation API

DashboardDataGrid:
- Add server-side pagination (serverTotal, serverTotalPages)
- Add hideFilter, bulkActions (renamed from children)
- Add header/rowCell snippet slots
- Support {#key} for forced re-render

/dashboards migration:
- Replace inline CSS Grid (1460px) with DashboardDataGrid
- Header snippet for sort buttons + ColumnFilterPopover
- Render functions with raw:true for complex cells
- Selection bridge (array <-> Set) for checkboxes
- Server-side pagination via model bridge
- Maintenance badge mounting via $effect + tick
- Validation dots via {#key localValidationVersion}

Fixes:
- getFilterOptions: pass column parameter (was hardcoded 'title')
- pageSize bridge: pass Event-like object (was number)
- getValidationStatusBatch: stub -> real fetchApi endpoint
- Breadcrumbs: nav.dashboard -> nav.dashboards
- Uncaught (in promise): add .catch() to async calls

Backend:
- New GET /status/batch endpoint for validation batch query

FileList rewrite (7 features):
1. Headers: text-sm font-semibold (was text-xs uppercase)
2. Column sorting (Name, Category, Size, Date)
3. Breadcrumbs with navigate-up button
4. Loading skeleton (animated rows)
5. Client-side pagination (20 per page)
6. Multi-select + bulk delete/download
7. Search by name/category

QA: build passes, 131 dashboard tests pass, 0 console errors
Pre-existing: 3 test failures unrelated (provider_config, api stub)

backend/src/api/routes/validation_tasks.py

@@ -121,7 +121,7 @@ async def create_task(
         extra={"src": "validation_tasks_routes"},
     )
     try:
-        result = service.create_task(payload)
+        result = await service.create_task(payload)
         scheduler.reload_validation_policy(result.id)
         return result
     except ValueError as e:
@@ -203,7 +203,7 @@ async def update_task(
         extra={"src": "validation_tasks_routes"},
     )
     try:
-        result = service.update_task(policy_id, payload)
+        result = await service.update_task(policy_id, payload)
         scheduler.reload_validation_policy(result.id)
         return result
     except ValueError as e:
@@ -259,7 +259,7 @@ async def toggle_task_status(
         extra={"src": "validation_tasks_routes"},
     )
     try:
-        result = service.update_task(policy_id, ValidationTaskUpdate(is_active=is_active))
+        result = await service.update_task(policy_id, ValidationTaskUpdate(is_active=is_active))
         if not is_active:
             scheduler.remove_validation_job(policy_id)
         else:
@@ -327,7 +327,7 @@ async def list_all_runs(
         extra={"src": "validation_tasks_routes"},
     )
     try:
-        total, raw_items = service.list_all_runs(
+        total, raw_items = await service.list_all_runs(
             status=status_filter, environment_id=environment_id, dashboard_id=dashboard_id,
             page=page, page_size=page_size,
         )
@@ -382,7 +382,7 @@ async def get_run_detail(
         extra={"src": "validation_tasks_routes"},
     )
     try:
-        result = service.get_run_detail(run_id)
+        result = await service.get_run_detail(run_id)
         return RunDetailResponse(
             run=_dict_to_run_response(result["run"]),
             records=result.get("records", []),
@@ -490,4 +490,70 @@ async def legacy_llm_report_redirect(
 # #endregion legacy_llm_report_redirect
 
 
+# #region get_validation_status_batch [C:3] [TYPE Function] [SEMANTICS validation,api,status,batch]
+# @BRIEF Batch-fetch latest validation status for multiple dashboard IDs.
+# @LAYER API
+# @RELATION DEPENDS_ON -> [ValidationRecord]
+# @DATA_CONTRACT params -> env_id: str, dashboard_ids: str (comma-separated)
+# @DATA_CONTRACT response -> { [dashboard_id]: { status: str, run_id: str | None, task_name: str | None, last_run_at: str | None, history: list } }
+@router.get("/status/batch")
+async def get_validation_status_batch(
+    env_id: str = Query(...),
+    dashboard_ids: str = Query(...),
+    db: Session = Depends(get_db),
+    current_user: User = Depends(get_current_user),
+):
+    """Return latest validation status for each requested dashboard ID."""
+    from ...models.llm import ValidationRecord, ValidationRun
+
+    ids = [id.strip() for id in dashboard_ids.split(",") if id.strip()]
+    if not ids:
+        return {}
+
+    # Get latest record per dashboard for this environment
+    from sqlalchemy import and_, desc
+    records = (
+        db.query(ValidationRecord)
+        .filter(
+            and_(
+                ValidationRecord.dashboard_id.in_(ids),
+                ValidationRecord.environment_id == env_id,
+            )
+        )
+        .order_by(desc(ValidationRecord.timestamp))
+        .all()
+    )
+
+    # Get run info for task names
+    run_ids = list({r.run_id for r in records if r.run_id})
+    runs = {}
+    if run_ids:
+        run_objs = db.query(ValidationRun).filter(ValidationRun.id.in_(run_ids)).all()
+        runs = {r.id: r for r in run_objs}
+
+    result: dict[str, dict] = {}
+    for rec in records:
+        if rec.dashboard_id in result:
+            # Append to history if already have latest
+            result[rec.dashboard_id]["history"].append({
+                "status": rec.status,
+                "last_run_at": rec.timestamp.isoformat() if rec.timestamp else None,
+                "task_name": runs.get(rec.run_id, {}).task_id if rec.run_id else None,
+                "run_id": rec.run_id,
+            })
+        else:
+            # First (latest) record sets the primary status
+            run = runs.get(rec.run_id)
+            result[rec.dashboard_id] = {
+                "status": rec.status,
+                "last_run_at": rec.timestamp.isoformat() if rec.timestamp else None,
+                "task_name": rec.task_id or (run.task_id if run else None),
+                "run_id": rec.run_id,
+                "history": [],
+            }
+
+    return result
+# #endregion get_validation_status_batch
+
+
 # #endregion ValidationTasksRoutes