`GroupBy.take()` and other missing functions

Currently, we can't do:
```kt
groupedDf
.take(10)
.concat()
```
to only concatenate the values of the first 10 groups. Instead, we'll have to convert to a normal DF first and convert back:
```kt
groupedDf
.toDataFrame().take(10).asGroupBy()
.concat()
```
The only row-based function that's available is `filter(GroupedRowFilter)` which can allow you to write `.filter { it.index() <= 10 }` but seems a bit odd.

Start by reading the GroupBy API around GroupBy.take(), filter(GroupedRowFilter), and the toDataFrame().asGroupBy() conversion shown in the issue. Compare the available grouped operations with the requested row-based behavior and clarify which other missing functions belong in scope. Done means the example can take the first 10 groups before concat(), with tests covering the resulting grouped operation.

Written by the indexing model from the issue text.