coordinate_matrix: Changed local thread size to 128 for better compatibility.

The use of the previous value 256 might be too large for some weaker hardware.
Since coordinate_matrix provides only poor performance anyway, we better
use a more portable value here without losing anything.