Update Advanced Search reindex logic after having issues in its own index

This issue probably should be handled together with Copy issues to new index. As we utilize Elasticsearch reindex API to handle reindex, we need to account for two indices when performing reindex tasks after we move issues off to its own index.

This change may need to be released together with Copy issues to new index. Or in order to iterate faster, we could also consider disabling the re-index feature temporarily if it's acceptable.

Epic order

1. Remove joins from the Elasticsearch query for project/group scoped issues search
2. De-normalize the issue permission data. <-- will include migration and can be done in multiple MRs
3. Remove joins from the Elasticsearch query for globally scoped issues search
4. Migration to copy issue data to new index
5. Migration to remove issue data from combined index **
6. Rake task changes to support new index **
7. Backend changes for supporting new indexes during Zero Downtime Re-indexing **
8. Admin UI changes to support new index (any time after new index created/data migrated to it)

** must be released in conjunction with the Migration to copy issue data to new index