PostgreSQL failover in GCP

1. Manually update repmgr priority of postgresql-02 in GCE to be 100: This will need to happen by hand. Entry is in schema repmgr_gitlab_cluster.repl_nodes in gitlab_repmgr database
2. Shutdown all postgresql servers in Azure.
3. Automatic promotion of postgresql-02 to master in GCE
4. Possibly restart the consul agent on pgbouncer, confirm the pgbouncer configuration.
5. Register postgresql-01 as part of the repmgr cluster
6. Manually update the priority of 03 / 04 so they are eligible to be masters