Support index snapshots for Disaster Recovery

For large Confluence environments, rebuilding the Lucene index can take several hours, which can be a necessary but disruptive task in the event that the existing index becomes corrupted. This improvement request is to track the ability for Confluence to take index snapshots for disaster recovery to mitigate the impact of such a scenario. This would further compliment the index recovery feature introduced in 5.8 for multi-node Confluence Data Center deployments.