fix(metrics): drop HTTP cache on metrics endpoint

Query-level cache in get_metrics_data is enough; view_cache added a second
layer that could serve stale responses after the warmed query cache updates.

Closes #1961

Signed-off-by: Alan Peixinho <alan.peixinho@profusion.mobi>

Author: alanpeixinho

backend/kernelCI_app/urls.py

@@ -8,7 +8,6 @@
 )
 
 from kernelCI_app import views
-from kernelCI_app.queries.notifications import METRICS_CACHE_TIMEOUT
 
 
 def view_cache(view, timeout: int = settings.CACHE_TIMEOUT):
@@ -180,9 +179,5 @@ def view_cache(view, timeout: int = settings.CACHE_TIMEOUT):
     path("proxy/", views.ProxyView.as_view(), name="proxyView"),
     path("origins/", views.OriginsView.as_view(), name="originsView"),
     path("tree-report/", views.TreeReport.as_view(), name="treeReportView"),
-    path(
-        "metrics/",
-        view_cache(views.MetricsView, timeout=METRICS_CACHE_TIMEOUT),
-        name="metricsView",
-    ),
+    path("metrics/", views.MetricsView.as_view(), name="metricsView"),
 ]