Bulk unpublish/archive harvested datasets

[dafeder]

Adds a new drush command, dkan:harvest:archive. Provided a harvest ID, it will archive all datasets connected to that harvest.

Note that connected items, like distributions, will not be touched. A separate PR will sync distributions workflow state to their parent datasets.

There is also a dkan:harvest:publish command that will do the inverse.

QA Steps

This harvest from demo.getdkan.com, so don't run these exact steps if you have the QA content.

• `drush dkan:harvest:register '{"identifier":"example","extract":{"type":"\Harvest\ETL\Extract\DataJson","uri":"https://demo.getdkan.org/data.json"},"transforms":[],"load":{"type":"\\Drupal\\harvest\\Load\\Dataset"}}'``
• drush dkan:harvest:run example
• Confirm that you have some published datasets.
• drush dkan:harvest:archive example
• Confirm that datasets are now unpublished.
• drush dkan:harvest:publish example
• Confirm that datasets are now published.

Note

I don't think there's a way to keep the method count under 20 on that one file CodeClimate is complaining about. I will be trying to break a lot of this stuff up in the coming weeks to refactor the metastore into some more manageable classes; lets revisit after that?

[dafeder]

Added a publish command. I think the redundant code complaint from codeclimate is ok for now; command classes are kind of their own thing (should clean up at some point through).

[janette]

Keep this branch until the next tag is cut, 2.13.2