glthread: determine global locking once every 64 batches to fix get_time perf

glthread: determine global locking once every 64 batches to fix get_time perf

This mitigates a large perf degradation when the clock source is HPET instead of TSC.
Just call get_time less frequently.

Fixes: 3ed141e9 - glthread: add a heuristic to stop locking global mutexes with multiple contexts
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/8910

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24168>