Add MPI Indexes to Improve Performance

[DanPaseltiner]

PULL REQUEST

Summary

This PR introduces indexing on the MPI in order to optimize performance of the new flattened schema. After profiling we have found that two categories of index improve performance.

1. Postgres does not automatically create indexes on foreign keys . Therefore introducing indexes on foreign key columns significantly improves performance of JOINS between tables.
2. Indexing based on the specific filter criteria for each block improves performance of identifying the records that belong in each block. We compared simply indexing each column versus being more specific e.g. using the first or last 4 characters of a given column. The later offered the best performance.

Related Issue

Fixes #CDCgov/phdi-azure#296

Checklist

• If this code affects the other scrum team, have they been notified? (In Slack, as reviewers, etc.)

[codecov]

Codecov Report

Merging #895 (588e633) into main (cd7efb5) will not change coverage.
The diff coverage is n/a.

@@           Coverage Diff           @@
##             main     #895   +/-   ##
=======================================
  Coverage   96.38%   96.38%           
=======================================
  Files          48       48           
  Lines        2821     2821           
=======================================
  Hits         2719     2719           
  Misses        102      102           

Flag	Coverage Δ
unit-tests	96.38% <ø> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

[m-goggins]

Looks good to me!