[C++][Compute] Make GroupBy optimizations work on Big Endian architecture

[asfimport]

This is related to ARROW-12010, which introduces optimized group by implementation.
On s390x (big endian architecture) the code hangs in unit tests (https://travis-ci.com/github/apache/arrow/jobs/506281680).
Currently this implementation is disabled on big endian and the alternative, earlier, implementation is used instead, but at some point we will stop maintaining both and switch to a single implementation, which requires fixing this issue.

Reporter: Michal Nowakiewicz / @michalursa

Related issues:

• [C++] Query engine umbrella issue (is a child of)

_{Note: This issue was originally created as ARROW-12830. Please see the migration documentation for further details.}

[github-actions]

This issue has been marked as stale because it has had no activity in the past 365 days. Please remove the stale label or comment below, or this issue will be closed in 14 days. If this improvement is still desired but has no current owner, please add the 'Status: needs champion' label.