Fix implicit conversions and unused variables in LAPACKE and TESTING (Reference-LAPACK PR 703)

Add a BLAS3-based triangular Sylvester equation solver (Reference-LAPACK PR 651)

Fix leading dimension check of eigen-/Schur vectors in LAPACKE (Reference-LAPACK PR 665)

Add NaN check functions for trapezoidal matrices to LAPACKE (Reference-LAPACK PR 738+742)

Fix workspace calculation in GEQRF/GERQF (Reference-LAPACK PR 638)

Fix workspace calculation in the left-looking variant of GEQRF (Reference-LAPACK PR690)

Add a trivial GEMMT implementation based on a looped GEMV

Fix cmake compilation of ReLAPACK and expose its INCLUDE_ALL option

Improve gcc arch option selecting for Neoverse cpus

Add [cz]scal microkernels for SKYLAKEX

Travis Ci: Increase the wait time for ppc jobs again

AzureCi: Limit cpu models in OSX_dynarch_cmake to keep it from running out of time

Add fallbacks to RaptorLake entry from PR3800

These are as similar to dscal_microk_skylakex-2.c as possible
for consistency.

Note that before this change SKYLAKEX+ uses generic C functions for
cscal/zscal via commit 2271c350 from #2610 (which is masked by
commit 086d87a30). However now #3799 disables FMAs (in turn enabled
by `-march=skylake-avx512`) in the plain C code which fixes excessive
LAPACK test failures more nicely.