A data engineering team is using Databricks to write data to a Delta table named "sensor_readings." They want to ensure that any duplicate records based on a specific key column are removed before writing to the table. Which code snippet should they use?