How do you handle potential data transformation errors during the parsing process in Nexthink pipelines?
