Checked job configuration and error-handler reporting to the shared monitoring project, including job-class tagging without sensitive arguments, while preserving existing retry behavior.
- What worked
- The existing reporter pattern clearly covered job failures alongside web errors with minimal configuration.
