Optimize cdot function for POWER10

This patch makes use of new POWER10 vector pair instructions for
loads and stores.

Fix building "generic" TRMM kernel with CMake

Add cblas_crotg, cblas_zrotg, cblas_csrot and cblas_zdrot

Fix compile-time setting of the GEMM buffer size for gmake builds

Added definitions for GEMM_PREFERED_SIZE and SWITCH_RATIO to the POWE…

Support NVIDIA HPC SDK on ARM64

Add CPUID information for Intel Rocket Lake

The CMake "TARGET_CORE" variables stores the "generic" target name in all lowercase letters, but gets compared to an all uppercase string, which results in the wrong TRMM kernel being selected.
This commit converts the TARGET_CORE to all uppercase before comparing its value to make sure case mismatches are not an issue in the future anymore.

Label the assembly part of the ARMV8 dynamic arch detection as volatile

rebase

Optimize swap function for POWER10

This patch makes use of new POWER10 vector pair instructions for
loads and stores.

Fix build on FreeBSD/powerpc64le

arm64: Fix nrm2 for input vectors with Inf

Fix double precision nrm2 kernels returning NaN when the
input vectors contain Inf/-Inf.

getarch.c: define OPENBLAS_SUPPORTED for riscv64

Expand the introductory paragraph of the README with links to netlib docs and linear algebra lecture videos

rebase

Define BLAS acronym in README

Temporarily revert to the old NRM2 kernels for ThunderX2/3 and NeoverseN1

Support NVIDIA HPC SDK on POWERPC