How to avoid column names like 'sum()' in aggregation in Spark/Scala?

Question

The aggregation

df.groupBy($"whatever").sum("A","B","C")

produces a DataFrame with column names like