[BUG] [Remote Store] [Auto Restore] Recover indices to green state after quorum loss recovery in remote backed clusters

[linuxpi]

Describe the bug

• During cluster state recovery, if cluster state is recovered from local disk, all shards would be initialized with ExistingStoreRecoverySource
• This leads to shards being in unassigned state and index in red. We manually trigger _remotestore/_restore for such indices after recovery to turn them green.
• For Remote backed shards ExistingStoreRecoverySource is not the best RecoverySource, they should be initialized with RemoteStoreRecoverySource once we have completed [Remote Store] Optimize remote store restore flow to not download all the files from remote if local disk already has data #9984
• This will allow us to skip the _remotestore/_restore call when cluster state is recovered from local disk after quorum loss

Expected behavior
After quorum loss recovery, if cluster state is recovered from local disk, remote backed indices should turn up green automatically