Restore ZEN SGEMM speed after #2361.

I partially reverted the changes in #2361 and I received the following
speed up on:
./xsl3blastst -R gemm -N 2048 2048 1 -a 5 1 1 1 1 1

AMD Ryzen 7 2700X (Zen+): 61400 to 63300 MFlops
AMD EPYC 7742 (Zen v2): 91400 to 94500 MFlops

These numbers are single-threaded performance.

kernel/x86_64/KERNEL.ZEN

@@ -30,8 +30,8 @@ DAXPYKERNEL = daxpy.c
 CAXPYKERNEL = caxpy.c
 ZAXPYKERNEL = zaxpy.c
 
-STRMMKERNEL    =  sgemm_kernel_8x4_haswell.c
-SGEMMKERNEL    =  sgemm_kernel_8x4_haswell.c
+STRMMKERNEL    =  sgemm_kernel_16x4_haswell.S
+SGEMMKERNEL    =  sgemm_kernel_16x4_haswell.S
 SGEMMINCOPY    =  ../generic/gemm_ncopy_8.c
 SGEMMITCOPY    =  ../generic/gemm_tcopy_8.c
 SGEMMONCOPY    =  ../generic/gemm_ncopy_4.c

param.h

@@ -666,7 +666,7 @@ USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 
 #else
 
-#define SGEMM_DEFAULT_P 320
+#define SGEMM_DEFAULT_P 768
 #define DGEMM_DEFAULT_P 512
 #define CGEMM_DEFAULT_P 256
 #define ZGEMM_DEFAULT_P 192