Improve output to show better the gaps and the suggested configurations

[david3107]

Issue

Currently the output is provided as multiple JSON objects and it differs per check. We fail the master check if at least one of the subchecks are not matching the policy. When the policy is not matched we mark the check with false

Improvement

While we can still fail the main master check, we can provide more details for the subchecks (or master checks) that are breaking the policy. For the ones that are not matching the policy also print a reference to the Github guidelines, our own threat model, MS DevOps Threat Matrix and SLSA.dev.