Follow-up to #2123. After the fix for the poison message loop, when a non-retriable error occurs during outbox event processing, the event is correctly skipped and the engine continues. However, the pipeline remains in its previous status in the database and the user has no visibility into the fact that the deployment failed.

This creates a situation where a pipeline appears to exist in the UI but was never actually deployed, with no indication of what went wrong.

Expected Behavior

• When a pipeline event is skipped due to a non-retriable error, the pipeline status should be updated to FAILED in the database with the error message.
• The frontend should display the failure state and the error details.
• The user should be able to trigger a re-deployment (e.g. via an update/save action, which fires a new outbox UPDATE event).

Scope

• Add a status and errorMessage field to the pipeline model.
• Update OutboxParentEventConsumer.consumeWithRetry() to set the pipeline status to FAILED when an event is skipped.
• Expose the failure state through the REST API so the frontend can display it.
• Frontend: show the error state and provide a retry action.