grt: fastroute: maze.cpp: Parallelized getOverflow3D with OpenMP

[glatosinski]

This PR:

• introduces usage of OpenMP to grt::FastRouteCore::getOverflow3D method to run overflow computation in parallel
• removes unused max_H_overflow and max_V_overflow computation

The comparison of speed (checked on 16 threads machine):

grt	master	PR
ariane	9:20 (min: 9:11, max: 9:36)	8:51 (min: 8:44, max: 8:47)
tinyRocket	0:25 (min: 0:25, max: 0:25)	0:22 (min: 0:22, max: 0:22)
ibex	0:43 (min: 0:42, max: 0:43)	0:42 (min: 0:42, max: 0:42)

[maliberty]

Note comments in #5215