opencl: add mul_mat_f32_f32_l4_lm and mul_mat_f16_f32_l4_lm
#4
| Job | Run time |
|---|---|
| 8s | |
| 8s |
mul_mat_f32_f32_l4_lm and mul_mat_f16_f32_l4_lm
#4
| Job | Run time |
|---|---|
| 8s | |
| 8s |