Increase stop_grace_period for Octavia controller services

The stop_grace_period for Octavia should be greater than the longest
task executed by Octavia.
The timeout of longest task in Octavia is currently set to 10 minutes
[0].

Updating the stop_grace_period from 5min to 10min30 will prevent
failures during updates. If a really long task is running Octavia (for
instance a failover during a network outage), it will timeout in Octavia
and Octavia will exit gracefully.

[0] https://review.opendev.org/c/openstack/octavia/+/637077

Change-Id: I05cd4a54c35479ca06580f541042948f69c7f532
(cherry picked from commit 1574403)

deployment/octavia/octavia-health-manager-container-puppet.yaml

@@ -157,7 +157,7 @@ outputs:
           map_merge:
             - octavia_health_manager:
                 start_order: 2
-                stop_grace_period: 300
+                stop_grace_period: 630
                 image: *octavia_health_manager_image
                 net: host
                 privileged: false

deployment/octavia/octavia-housekeeping-container-puppet.yaml

@@ -127,7 +127,7 @@ outputs:
         step_5:
           octavia_housekeeping:
             start_order: 2
-            stop_grace_period: 300
+            stop_grace_period: 630
             image: *octavia_housekeeping_image
             net: host
             privileged: false

deployment/octavia/octavia-worker-container-puppet.yaml

@@ -115,7 +115,7 @@ outputs:
         step_5:
           octavia_worker:
             start_order: 2
-            stop_grace_period: 300
+            stop_grace_period: 630
             image: *octavia_worker_image
             net: host
             privileged: false