with parsers, either built in or hand crafted jelly script, you can throw a lot of different data into the warnings buckets.

it would be nice to be able to select from either category or type to indicate which portions of those data you want on a dashboard view.

CONOPS:

lots of little quick jobs that each have multiple checkers with associated parsers contributing to warnings, so developers get a soup to nuts validation of their commit

best practice review wants to see sorting/trending of each of those warnings separately to look for hot spots, etc.

in this case its not realistic to split the jobs into tool name filterable pieces (too many jobs, and warnings don't roll up if jobs are connected in a stream)
its also not realistic (no matter how desirable) that those individual checkers will get their own plugins and columns in analysis-code in any reasonable amount of time.