"Add Filter" Performance Issue

[jqnatividad]

CKAN Version if known (or site URL)

2.5.7

Please describe the expected behaviour

When using "Add Filter" in a datatables resource view, the "Add Filter" should load-paginate the distinct values of a given column in a performant manner, leveraging the user-keyed value to smartly search/autocomplete the dataset resource.

Please describe the actual behaviour

if you go to

https://data.boston.gov/dataset/311-service-requests/resource/2968e2c0-d479-49ba-a884-4ef523ada3c0

and do "Add Filter", select "CASE_TITLE" and type in "Request for Pothole Repair", it takes a long time
the freetext search on the right, returns 55K rows in less than 5 secs for the same thing
It seems the root cause is that the "Add Filter" loads the distinct values for CASE_TITLE and there are about 13k distinct vals given its a freetext field... any tweaks we can do in the "Add Filter" functionality to make it smarter - i.e. as the use types, it only loads vals starting with the keyed value?

What steps can be taken to reproduce the issue?