Make enable-shard-allocation persistent?

[geofffranks]

Right now, enable-shard-allocation is an errand that seems to need running after every elasticsearch restart, in order to keep a healthy cluster.

Would it be possible to modify it so that it updates via a persist flag, rather than transient, and have this run as a post-start action on the master node? Or would that cause undesirable effects in elasticsearch over the long-term?

[Infra-Red]

Actually this changes has been already implemented, please check #39.