This changes the kernels to pack full SVE vectors and reduces the
overall complexity of the inner GEMM loop.

ARM64: Enable bfloat16 kernels by default

This re-uses the existing NEOVERSEN2 8x4 `sbgemm` kernel to implement `bgemm`.

- Adds bgemv T based off of sbgemv T kernel
- Adds bgemv N which is slightly alterated to not use Y as an
accumulator due to the output being bf16 which results in loss of
precision
- Enables BGEMM_GEMV_FORWARD to proxy BGEMM to BGEMV with new kernels

SME1 based direct kernel (with alpha and beta) for cblas_sgemm level 3

This also improves the testing and generic kernel by re-using the BF16
conversion functions.

Built on top of https://github.com/OpenMathLib/OpenBLAS/pull/5357 and derived from https://github.com/OpenMathLib/OpenBLAS/pull/5287

Co-authored-by: Ye Tao <ye.tao@arm.com>

Add support for Ampere AmpereOne processors

kernel/arm64: fixed cscal and zscal

- Calculate predicate outside the loop
- Divide matrix in blocks of 3

* Fix ARMV9SME target and add support_sme1 code for MacOS
* make sgemm_direct unconditionally available on all arm64
* build a (dummy) sgemm_direct kernel on all arm64





* Update dynamic_arm64.c

Add vector registers to clobber list to prevent compiler optimization.

Use SVE kernel for S/DGEMVN for SVE machines

Improve performance for SGEMVN on NEONVERSEN1

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

SME based SGEMMDIRECT kernel uses the vector registers (z) and adding
    clobber list informs compiler not to optimize these registers.

Optimize aarch64 sgemm_ncopy

replace customize bf16_to_fp32 with arm neon vcvtah_f32_bf16

Optimize gemv_n_sve kernel

Add dispatch of SBGEMVNKERNEL for NEOVERSEN2 and NEOVERSEV2

This improves performance for sbgemv_t by up to 100x on NEOVERSEV1.
The geometric mean speedup is ~61x for M=N=[2,512].