ppbs_descr

Pivot Partitioning by Scanning (PPbS) rearranges the elements in an array A around a given
pivot element in place, much like the partitioning procedure used in the Quicksort algorithm,
e.g. the one shown in the famous CLRS book.
There are some differences though: whereas the algorithm from CLRS partitions the array into two parts
(elements <= pivot and elements > pivot), given a pivot element which always is the last element in
the array, PPbS partitions the array into three parts (<, ==, and > than the pivot) with a freely chosable
pivot. PPbS, however, suffers from nested loops, so its time complexity is worse.

PPbS returns two pointers, m1 and m2, that satisfy the following conditions:
   1. There are m1 many elements that have a value < pivot
   2. There are (m2 - m1) many elements that have a value == pivot
   3. There are (n - m2) many elements that have a value > pivot, where n == A.length.