We should more clearly separate the C-Heap version from the arena / resource area allocation.

Advantages:
- If they are separate subclasses, then we have to do less verification when assigning, copying, allocating etc: fewer mistakes can happen.
- We can establish a clear separation between: take care of allocation/deallocation yourself (C-Heap) vs. abandon memory and the elements still in the array (arena / resource area).
- We should not have multiple implementations of the same thing (C-Heap backed array).