Add heapq.MinHeap and heapq.MaxHeap, an object-oriented interface to the existing heapq functions.

Both classes wrap a list and expose the heap operations as methods, so you no longer have to pass the list around to the module-level functions:

>>> from heapq import MinHeap
>>> heap = MinHeap([3, 1, 2])
>>> heap.push(0)
>>> heap.pop()
0
>>> len(heap)
3

The classes provide push, pop, pushpop, replace, nsmallest and nlargest, plus __len__, __bool__ and __repr__. MaxHeap behaves as a max-heap, using the *_max variants under the hood.

The implementation is pure Python and delegates to the C-accelerated module functions, so there is no performance regression.

Includes documentation, a What's New entry and tests.

• Issue: Add MinHeap and MaxHeap classes to heapq (object-oriented interface) #153357