<execution>: Parallelize more algorithms

Some strategies that can be used:

Just call parallel transform:

• replace_copy
• replace_copy_if

Scans:

• copy_if
• partition_copy
• remove_copy
• remove_copy_if
• unique
• unique_copy

Same as serial nth_element but call the parallel partition op for large N:

• nth_element

Predicate tests (like all_of):

• lexicographical_compare

Summary statistics (like find / find_end):

• min_element
• max_element
• minmax_element

Divide and conquer:

• inplace_merge
• stable_partition

Divide range1 into chunks, binary search to find matching range2 chunks, scan:

• merge
• set_symmetric_difference
• set_union

Other:

• includes

Start with the algorithm groups and parallelization strategies listed in the issue, then inspect how the project currently handles the corresponding serial algorithms. Done means the listed algorithms have been parallelized using the proposed approaches and their behavior is covered by the project's validation process.