Improve rules creation using GeoFence batch

[etj]

The creation of the Rules in GeoFence should be as quick as possibile, since it may lead to excessive waiting time and, in some cases, to the failure of the logic related to guardian/geofence sync.

At the moment, in order to create a rule in GeoFence, 3 REST API calls are invoked:

• get the number of rules
• get the last rule (in order to read its priority)
• create the new rule after the last existing rule

(btw, the priority computation at the moment is wrong, since it reuses the last used priority)

In order to improve the performance of rule creation, we can leverage on the use of GeoFence batch;
refs:

• Jira - implementation in GeoServer: https://osgeo-org.atlassian.net/browse/GEOS-10507
• Original GeoFence REST doc: https://github.com/geoserver/geofence/wiki/GeoFence-REST-API#batch-operations
  Also, I noted that the syntax for the batch command implemented in GeoServer does not match the original (standalone) one:
  https://docs.geoserver.org/latest/en/user/extensions/geofence-server/rest-batch-op.html
  Of course at the moment we'll follow the syntax for the embedded GeoFence, but the API will need to be aligned on GeoServer.

Instead of issuing REST calls for creating a single rule, we can

• compute the rule priority using the existing 2 REST calls (pls note that this is a shortcoming due to the bad implementation of the rule insertion in the embedded geofence, since the original GeoFence API allowed more flexibility in the rule piority positioning
• create a Batch with all the needed rules
• add the rules in GeoFence in a single transactional call