Wrong number of elements is copied with async_work_group_copy in kernel1

This instruction in "debugfastergrad" branch copies GENOTYPE_LENGTH_IN_GLOBMEM elements from global memory into the local array genotype[ACTUAL_GENOTYPE_LENGTH].

However, such array is defined with a size smaller than what is copied into it. Therefore, the aforementioned instruction should copy ACTUAL_GENOTYPE_LENGTH instead of GENOTYPE_LENGTH_IN_GLOBMEM.