Table level bifurcation required in pull query metrics

Hi Team

I have a single ksqlDB (0.14 community) cluster where i have 2 separate aggregate tables(T1 and T2) from which we are serving pull queries at the same time.

Sample pull queries are as follows :-

1. select * from T1 where uid = 12 ;
2. select * from T2 where pid = 13 ;

Here we want to monitor pull query performance of both these tables separately to get clear idea of performance at individual table level. As of now it seems pull query metric doesn't give this bifurcation.

Start by reviewing the existing pull query metrics for ksqlDB 0.14 and how queries against tables T1 and T2 are represented. Determine how metrics could distinguish the two tables, then define the expected separate performance measurements for the sample queries and add coverage for that behavior.

Written by the indexing model from the issue text.