Create SHERLOCK-PIPEline as a service

[martindevora]

We could have a long-living script able to work as a service: Every minute it could poll the given user defined YAML or CSV files and keep a copy of them as a queue. It could be nice to be able to define priorities (for instance, from 0 to 3). Then, the service would launch them sequentially taking into account the priorities.

The pipeline would provide an add method for the users to queue the new requests as YAML or CSV files.

Ideally, the service would be able to notify the users as soon as their request had already been processed.