Improve GitLab Geo Planned Failover and Disaster Recovery documentation

Improve GitLab Geo Planned Failover documentation https://docs.gitlab.com/ee/gitlab-geo/planned-failover.html

• Merge this doc into Step 1 of https://docs.gitlab.com/ee/gitlab-geo/disaster-recovery.html
  • And say "Attempt to complete as much of this as possible even if this is an unplanned recovery"
• In the Block primary traffic step, add a step to verify the GitLab UI is blocked. E.g. Verify that you cannot visit the primary in browser
• In the Block primary traffic step, add a step to verify pushes are blocked. E.g. Verify that pushing a commit fails over HTTP
• In the Block primary traffic step, add a step to verify pushes are blocked. E.g. Verify that pushing a commit fails over SSH if Git over SSH is enabled
• In the Allow replication to finish as much as possible step, specify exactly which queues are of most interest This is addressed sufficiently in https://gitlab.com/gitlab-org/gitlab-ee/merge_requests/4920
• Add "FYI remember to remove broadcast message" after failover is complete
• Suggest lowering TTL in the DR doc after preparing for planned failover (this advice is useful for unplanned failovers as well)
• Reassure that changing the external url won’t prevent access via the secondary url, as long as the secondary DNS records are correct.