Fix CBLAS_?GEMMT filling in the wrong triangle for Row-Major

TST,BUG: Explicitly allow running tests multiple times

Update Android NDK install path for M1/armv7 crossbuild on CirrusCI

CGEMM & ZGEMM using C code, Power only, P10 only.

fix: add missing NO_AFFINITY checks

Add preliminary cpu autodetection for Zen5/5c

Update names of Loongarch64 targets in cmake cross-building

Fix potential inaccuracy in multithreaded level3 related to SWITCH_RATIO

Handle inf/nan in ppc440 s/dscal

Without failures due to existing files

Fix SIGILL/SIGSEGV in PPCG4 SGEMM and fix NAN handling in PPCG4 SSCAL/DSCAL

ARM64: Enable SMALL_MATRIX_OPT when compiling with CMake

Support LoongArch64 compilation with LLVM

LoongArch64: Enable cmake cross-compilation

Declare the input array in CBLAS_?GEADD as const in cblas.h

fix placement of -fopenmp in the pkgconfig file

LoongArch64: Rename core

Use microarchitecture name instead of meaningless strings to name the core,
the legacy core is still retained.
1. Rename LOONGSONGENERIC to LA64_GENERIC
2. Rename LOONGSON3R5 to LA464
3. Rename LOONGSON2K1000 to LA264

Strip any mtune option from FFLAGS is the compiler is flang-new