Fix blob id too long

Issue #5870

• Adds support for blobs with long paths so they can be indexed by Elasticsearch
• the ID is currently created by combining project_id and blob path, Elasticsearch has a hard limit of 512 length
• any generated BlobIDs over 512 will have the path hashed first, new format _ (continuing to prefix the projectId will avoid any potential duplicate hashes across projects)

Note: The solution to have multiple ways of generating blob ids is not ideal, but it does allow for items currently in the index to function with any changes and allows items previously failing to be indexed.

This GitLab change will be required to use a new version of the indexer and will add tests verifying the fix