Add UI for observing outcome when deploying policies in "dry run" mode

[nscuro]

As part of #957, we identified that it would be good to have a "dry run" mode for policies. This mode would evaluate policies, but not apply their associated action.

Initially, this functionality is implemented using application logs. However, long-term, we can't expect users to observe logs. We should have a UI integration that provides this information instead.

The granularity of the information being displayed is still subject for discussion. But generally, the following would be nice to have:

• Number of affected projects
• Number of affected components
• Number of affected findings
  • Potentially with additional aggregates, as in "avg. findings affected per component / project"

[skhokhlov]

I'm thinking maybe DT shouldn't apply new bundle of policies automatically after pulling, but wait for review? I mean firstly apply them in dry-run mode and then wait for approve

[VinodAnandan]

It's a good idea, but keeping it as the default behaviour while still allowing overrides makes it useful for multiple scenarios. Some may prefer to maintain a strict review and approval process solely within Git, without using multiple systems (Git and DT) for policy approval.

[skhokhlov]

Right, this may be configurable