Use 2 columns from PySpark data frame as lookup to replace all other columns

Question

I have a data set that is a supervisory hierarchy and the first two columns are id and name and the following columns are level 1, l