swh/production: Increase counters rpc timeout

The refresh_history endpoint takes more than 10s to
retrieve the data from prometheus and compute the values.
The cronjob regularly failed probably until enough
data is in the thanos cache.

Related to swh/infra/sysadm-environment#5387

swh/values/production/swh.yaml

@@ -2048,7 +2048,7 @@ counters:
     gunicorn:
       workers: 2
       threads: 1
-      timeout: 10
+      timeout: 30
     # autoScaling:
     #   minReplicaCount: 2
     #   maxReplicaCount: 4