fix(resource metrics): throttle resource metrics handling error message

If for whatever reason handling a "hot" metric such as `matched` breaks, it's likely that
logs will be flooded with identical messages.

apps/emqx_conf/src/emqx_conf_schema.erl

@@ -84,6 +84,7 @@
     dropped_msg_due_to_mqueue_is_full,
     external_broker_crashed,
     failed_to_retain_message,
+    handle_resource_metrics_failed,
     socket_receive_paused_by_rate_limit,
     unrecoverable_resource_error
 ]).

apps/emqx_resource/src/emqx_resource_metrics.erl

@@ -130,10 +130,11 @@ handle_telemetry_event(
             %% We catch errors to avoid detaching the telemetry handler function.
             %% When restarting a resource while it's under load, there might be transient
             %% failures while the metrics are not yet created.
-            ?SLOG(
+            ?SLOG_THROTTLE(
                 warning,
+                ID,
                 #{
-                    msg => "handle_resource_metrics_failed",
+                    msg => handle_resource_metrics_failed,
                     hint => "transient failures may occur when restarting a resource",
                     kind => Kind,
                     reason => Reason,
@@ -158,10 +159,11 @@ handle_telemetry_event(
             %% We catch errors to avoid detaching the telemetry handler function.
             %% When restarting a resource while it's under load, there might be transient
             %% failures while the metrics are not yet created.
-            ?SLOG(
+            ?SLOG_THROTTLE(
                 warning,
+                ID,
                 #{
-                    msg => "handle_resource_metrics_failed",
+                    msg => handle_resource_metrics_failed,
                     hint => "transient failures may occur when restarting a resource",
                     kind => Kind,
                     reason => Reason,