[Database] DB schema and migrations to support filtering of vulnerabilities by identifier

As per the requirement and the MVC scope discussed in #432419 (closed) we need to have identifiers stored in vulnerabilitiy_reads table to support filtering of vulnerabilities by identifiers on the vulnerability report page.

In this issue we track the DB related work to implement the feature.

Implementation details

Abandoned old plan:

Reason: #460080 (comment 2007295428)

• backend database Include identifiers column of array text type in vulnerability_reads.
• backend database Ingest new vulnerability identifiers into the array column in vulnerability_reads.
• backend database Data backfill migration to populate the already existing identifiers to vulnerability_reads.

New plan:

With the requirement to use identifier.name in the vulnerability_reads array column the revised implementation looks like:

• backend database Abandon the already running databackfill migration from the old implementation plan above. Process around this is available in !160594 (merged)
• backend database Disable the existing ingestion logic and ignore the old column identifier_external_ids. Process to follow https://docs.gitlab.com/ee/development/database/avoiding_downtime_in_migrations.html#dropping-columns
• backend database Include new identifier_names column of array text type in vulnerability_reads. Renaming older column is not possible as the old column has a default value []. See: link.
• backend database Revise Ingestion logic to populate identifiers.name into the array column in vulnerability_reads.identifier_names.
• backend database Data backfill migration (BBM) to populate the already existing identifiers to vulnerability_reads on the new identifier_names column.