optimise dot using thread throttling for NEOVERSE V1

Use SVE kernel for S/DGEMVN for SVE machines

Fix: Potential out-of-bounds read in non-transposed [SD]GEMV kernels for A64FX and Neoverse V1.

remove non-vectorized SGEMV transpose reduce path for POWER8, restoring optimizations frpm PR4880

Fix and reinstate the Cooper Lake/Sapphire Rapids microkernel for non-transpose SBGEMV

Explicit unaligned vector load/stores in PPC64LE GEMV kernels

Wrong output from getarch on Apple M4

Follow-up to #5233, fixing "Argument list too long"

Fix "Argument list too long" compilation error for Intel macOS

Temporarily disable the Cooper Lake/Sapphire Rapids microkernel for non-transpose SBGEMV

Bump xuantie qemu for c910v

Signed-off-by: Han Gao <rabenda.cn@gmail.com>

Improve performance for SGEMVN on NEONVERSEN1

Update and amend parameters for Neoverse cpus in CMake crossbuilds

Further performance improvements to non-transposed [SD]GEMV kernels for A64FX and Neoverse V1.

Add AArch64-optimized SYMV kernels

lapacke_mangling.h is no longer generated, so don't delete on make clean

Use SVE kernel for S/DGEMVT for SVE machines

Optimizing the Implementation of GEMV on the RISC-V V Extension

test_potrs.c: do not use GCC pragma on darwin-aarch64