Using `vulnerability_reads` table to speed up API responses

Why are we doing this work

As we create a new table called vulnerability_reads to query vulnerability data efficiently without the necessity of joining different tables, we need to change our logic to generate API responses based on this new table to improve the performance of our software.

Relevant links

• The issue to create the data model: #335326 (closed)

Implementation plan

• backend Implement a new finder class to utilize the vulnerability_reads table to find vulnerability records
• backend Create a new feature flag to use either the new or old finder class to fetch the vulnerability records