AVX512 CGEMM & ZGEMM kernels

96-99% 1-thread performance of MKL2018

rebase

Optimize SGEMM on SKYLAKEX CPUs

Disable three-operand DCBT on PPC970 regardless of operating system

Disable IDAMIN/MAX and IZAMIN/MAX optimizations on big-endian POWER8

Remove obsolete workarounds for IOS on ARMV8

There seems to be no advantage to the three-operand form used in the earliest GotoBLAS kernels, and it causes compilation problems  on other than the previously special-cased platforms as well

resync with upstream

Fixed a minor cmake problem, occuring when DYNAMIC_ARCH=ON and CMAKE_C_FLAGS was empty

Add support for NetBSD by adding it to the existing xBSD conditionals

Both forms are seen on NetBSD (#2288)

Improve support for g95 and non-GNU ld

Fix netlib 3.7/3.8 function enumeration for linktest

Optimize RPCC function on ARM64

Fixup x86 cpuid changes from #2283

Auto-add "-fno-second-underscore" option to make LAPACKE compile (as it calls LAPACK functions that may have gotten a second underscore added otherwise). Also support -R for rpath when parsing compiler directives in f_check

to allow linktest to pass with a compiler that adds a second underscore to such names

as I find conflicting information on accessibility from non-priviledged processes

* Fix source typo

Found via `codespell -q 3 -L amin,als,ba,dum,mone,nd,nto,orign -S Changelog.txt,./lapack*`

* Remove beta-thread function per request

#2283 had inadvertently removed Goldmont+, and cpuid was reporting a mix of Core2 and Pentium2 for some QEMU configurations

resync with upstream

AVX512 DGEMM kernel

90% MKL 1-thread performance.

Support QEMU virtual cpu in 64bit mode as CORE2 or BARCELONA