df.maxFor { age and name } is not allowed, even though the result would be a DataRow containing the maximum value for each column independently.
df.maxFor { (age and name).asComparable() } works fine, so it's just the type limitation.
It's probably fine to change the type to C : Comparable<*>? for these particular functions, as C is never used in a reified way.
This also applies to medianFor, and percentileFor