duplicate a column in pyspark data frame