As part of our effort to enhance the analytics for Risk Analysis we want to consider recent data persisted via TRT-1702.  If we detect a potentially risky regression we want to see if that same test failure is occurring in other PRs. If so this lowers the potential that the regression is within the current PR and we can consider persisting the data but suppressing the comment.

For this story we want to query persisted potential regressions detected by risk analysis in big query within a recent timeframe, possibly last 24 hours.  A potential place to insert logic is in pr_commenting_processor.

We want to detect when a similar failure is found in PRs not associated with our current PR (some risk here if we have multiple copies of the PR) when we analyze across repos we can place additional emphasis on similar failures across repos.  When we do so we can either add information to the comment that this is occuring outside of the current PR as well or just suppress the comment for that test.  Need to discuss among the team on the best approach.