Extend Security::FindingTokenStatus model to support Secret Detection features

Why are we doing this work?

Based on this discussion we must persist raw_source_code_extract somewhere in the database so that we do not introduce a regression.

In [SPIKE] Figure out what to do with `raw_source_... (#565033 - closed) we agreed that the best solution is to extend functionality of Security::FindingTokenStatus model and create Security::FindingTokenStatus records during Vulnerability ingestion.

In order to extend the Security::FindingTokenStatus model we have to add raw_source_code_extract column to it.

Implementation plan

1. Create a regular migration to add raw_source_code_extract column to security_finding_token_statuses with a text limit