Index and update every job to an ElasticSearch document

[mmenozzi]

To be able to fetch job data easily and quickly in a web console, every job should be indexed into an ElasticSearch document. Alongside with job data also a kind of "job state" should be indexed as well (for example the state that the job has in the Beanstalk queue).

• Index new job just when produced
• Update already indexed job when reserved by worker
• Update already indexed job when successfully worked by worker
• Update already indexed job when worked with error by worker
• Remove usage of temporary repo https://github.com/mmenozzi/amp-elasticsearch
• Check if a new Job's uuid is already indexed onto ElasticSearch and throw it's the case

[mmenozzi]

Merged in master. Closing.