Build collection page in catalog-next

[FuhuXia]

User Story

In order to list all datasets of the same isPartOf value, data.gov team wants have a collection page.

Acceptance Criteria

[ACs should be clearly demoable/verifiable whenever possible. Try specifying them using BDD.]

• GIVEN there are such three datasets from the same harvest source (id=my_sid)
  AND one dataset with identifier=my_pid, name=my_parent
  AND one dataset with identifier=my_cid1, name=my_child1, isPartOf=my_pid
  AND one dataset with identifier=my_cid2, name=my_child2, isPartOf=my_pid
  WHEN its collection page is loaded
  THEN datasets my_child1 and my_child2 are listed
  AND user can search within this collection.

Background

This can be done with Solr query using isPartOf and harvest_source_id as search terms, as in
https://catalog-next-dev-datagov.app.cloud.gov/dataset/?q=isPartOf:pid+AND+harvest_source_id:sid. But the resulted page has the search box prefilled with search terms that users do not understand, therefore users wont be able to perform their own search. We want a clean page starting with empty in the search box, and make it obvious that the current listing is a collection, same as how collection is represented in the current catalog.data.gov, such as this example.

Security Considerations (required)

[Any security concerns that might be implicated in the change. "None" is OK, just be explicit here!]

Sketch

[Notes or a checklist reflecting our understanding of the selected approach]
Instead of using raw SOLR query as q=isPartOf:pid+AND+harvest_source_id:sid, we can add a new term collection_info=sid+pid, CKAN will grab the collection_info value and split it into harvest_source_id and isPartOf to be used in the python code to find all datasets that belong to the collection.

[FuhuXia]

It turns out /dataset/?isPartOf=#&harvest_source_id=# is clean enough as in this sample page, just need to hide the two blocks inthe circle. Thanks to @jbrown-xentity.