AI-GCS: Execute at-scale evaluation and tuning campaigns of end-to-end solution

[pombredanne]

To evaluate, refine and tune how the code search works, we should execute tests, evaluations, bug fixing and tuning campaigns at a large enough scale for the whole end-to-end solution using the test and reference datasets collected before.

The results should be:

• Publish automation/test scripts to automate these campaigns.
  • Create evaluation reports.
• Apply tuning and updates from evaluation campaigns
• Then, make a new release of MatchCode.io/PurlDB/ScanCode.io as needed