Improve the performance of dasum and sasum when SMP is defined

Add reproducer test for crash after fork

Don't overwrite blas_thread_buffer if already set

See #2993 for an analysis

After a fork it is possible that blas_thread_buffer has already
allocated memory buffers: goto_set_num_threads does allocate those
already and it may be called by num_cpu_avail in case the OpenBLAS
NUM_THREADS differ from the OMP num threads.
This leads to a memory leak which can cause subsequent execution of BLAS
kernels to fail.

Fixes #2993

Fix sum optimize issues

Optimize the performance of rot by using universal intrinsics

Fix missing -mfma compiler flag in cmake builds without DYNAMIC_ARCH

rebase

Merge branch 'openblas-open-910' of git://github.com/damonyu1989/OpenBLAS into damonyu1989-openblas-open-910

Improve the performance of rot by using AVX512 and AVX2 intrinsic

Fix missing AVX2 and FMA3 capabilities in FORCE_target mode

Optimize sdot/ddot for POWER10

Fix handling of cpu capability flags in DYNAMIC_ARCH builds

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

rebase

Fix macro name used in ifdef for POWERPC/PGI

Ensure that EXPRECISION is disabled for DYNAMIC_ARCH with TARGET=GENERIC and fix CMAKE DYNAMIC_ARCH builds

POWER10:  Change dgemm unroll factors