This change adds the unfollow action for CCR follower indices.

This is needed for the shrink action in case an index is a follower index.
This will give the follower index the opportunity to fully catch up with
the leader index, pause index following and unfollow the leader index.
After this the shrink action can safely perform the ilm shrink.

The unfollow action needs to be added to the hot phase and acts as
barrier for going to the next phase (warm or delete phases), so that
follower indices are being unfollowed properly before indices are expected
to go in read-only mode. This allows the force merge action to execute
its steps safely.

The unfollow action has three steps:
* `wait-for-indexing-complete` step: waits for the index in question
  to get the `index.lifecycle.indexing_complete` setting be set to `true`
* `wait-for-follow-shard-tasks` step: waits for all the shard follow tasks
  for the index being handled to report that the leader shard global checkpoint
  is equal to the follower shard global checkpoint.
* `unfollow-index` step: actually performs the unfollow. This consists
  out of multiple operations being executed on the index being handled:
  pause index following, close index, unfollow and open index. (a follower
  index can only be unfollowed when it is closed, because the underlying
  engine is changed)

In the case of the last two steps, if the index in being handled is
a regular index then the steps acts as a no-op.

Relates to elastic#34648

changed WaitForIndexingComplete step to always ignore non follower indices.

This aligns WaitForIndexingCompleteStep with naming conventions

The shard follow tasks continuesly poll changes from leader index,
in case no new writes arrive in the leader shards, then the leader
shards wait up to the defined read_poll_timeout before returning an
empty result. The result also includes information about whether
the index settings need to be synced.

Because in this test no data was indexed it took a full minute for ccr
to realize that the followe index settings needed to be updated and
caused the test to fail due to an assert busy statement to time out.

open follower index operation from UnfollowFollowerIndexStep
to the following steps respectively:
* PauseFollowerIndexStep
* CloseFollowerIndexStep
* OpenFollowerIndexStep

The UnfollowAction now exists of the following steps:
* WaitForIndexingCompleteStep
* WaitForFollowShardStep
* PauseFollowerIndexStep
* CloseFollowerIndexStep
* UnfollowFollowerIndexStep
* OpenFollowerIndexStep

so that after unfollow is done then the unfollowed index is ready
to handle requests.

because after the unfollow step it is no longer a follower index.

This is necessary, at least in the Hot phase, so that the priority will
be set before waiting for the index to unfollow, which is much more
likely what the user expects.