help wantedkind/featurepriority/backlog
Description
We use trivy server to scan images from both CI pipelines and a Kubernetes operator. Our log infrastructure (Splunk) is much more fond of logs in machine-readable format (JSON), so it would be nice if it were possible to configure the log format when bootstrapping trivy server.
It seems like trivy uses https://github.com/uber-go/zap as the logging backend, supporting configuration, and JSON logging. A potential challenge could be that it seems hard to integrate Zap configuration with Viper/Cobra, ref. https://github.com/uber-go/zap/issues/542.