Travis CI: Add a DYNAMIC_ARCH build on Neoverse using older gcc8

Remove SVE from Arm(R) Neoverse(TM) N1 CPU in Makefile

I incorrectly added `+sve` to the Neoverse(TM) N1 CPUs GCC parameters,
which doesn't support SVE - this results in failed builds when using a
compiler that doesn't support `-mtune=neoverse-n1` which appears to hide
the mistake.

dscal: use ymm registers in Haswell microkernel

Allocate work array in LAPACKE ?TGSEN when ijob is zero (Reference-LAPACK PR 733)

Check for NaN in ?GECON (Reference-LAPACK PR765)

Fix uninitialized M in quick return from D/SLARRD (Reference-LAPACK PR769)

Set SWITCH_RATIO for Arm(R) Neoverse(TM) V1 CPUs

scal benchmark: eliminate y, move init/timing out of loop

Add SVE implementation for sdot/ddot

Using 256-bit registers in dscal makes this microkernel consistent with
cscal and zscal, and generally doubles performance if the vector fits in
L1 cache.

This adds an SVE implementation to sdot/ddot when available, falling back to the previous Advanced SIMD kernel where there's no SVE implementation for the kernel.

All the targets were essentially treating `dot_thunderx2t99.c` as the Advanced SIMD implementation so I've renamed it to better fit with the feature detection.

Remove unnecessary instructions from Advanced SIMD dot

From testing this yields better results than the default of `2`.

Removing y avoids cache effects (if y is the size of the L1 cache, the
main array x is removed from it).
Moving init and timing out of the loop makes the scal benchmark behave like
the gemm benchmark, and allows higher accuracy for smaller test cases since
the loop overhead is much smaller than the timing overhead.

Example:
OPENBLAS_LOOPS=10000 ./dscal.goto 1024 8192 1024
on AMD Zen2 (7532) with 32k (4k doubles) L1 cache per core.

Before
From : 1024  To : 8192 Step = 1024 Inc_x = 1 Inc_y = 1 Loops = 10000
   SIZE       Flops
   1024 :     5627.08 MFlops   0.000000 sec
   2048 :     5907.34 MFlops   0.000000 sec
   3072 :     5553.30 MFlops   0.000001 sec
   4096 :     5446.38 MFlops   0.000001 sec
   5120 :     5504.61 MFlops   0.000001 sec
   6144 :     5501.80 MFlops   0.000001 sec
   7168 :     5547.43 MFlops   0.000001 sec
   8192 :     5548.46 MFlops   0.000001 sec

After
From : 1024  To : 8192 Step = 1024 Inc_x = 1 Inc_y = 1 Loops = 10000
   SIZE       Flops
   1024 :     6310.28 MFlops   0.000000 sec
   2048 :     6396.29 MFlops   0.000000 sec
   3072 :     6439.14 MFlops   0.000000 sec
   4096 :     6327.14 MFlops   0.000001 sec
   5120 :     5628.24 MFlops   0.000001 sec
   6144 :     5616.41 MFlops   0.000001 sec
   7168 :     5553.13 MFlops   0.000001 sec
   8192 :     5600.88 MFlops   0.000001 sec

We can see the L1->L2 switchover point is now where it should be, and the
number of flops for L1 is more accurate.

The existing kernel was issuing extra instructions to organise the arguments into the same registers they would usually be in and similarly to put the result into the appropriate register.

This has an impact on smaller sized dots and seemed like a quick fix

Fix SLATRS3 and CLATRS3 tests in TESTING/LIN (Reference-LAPACK PRs 755+764)

Fix typo in EIG tests and spurious return in lapacke_?tz_trans utility (Reference-LAPACK PR760)

Fix array dimesion in complex SYL01 test (Reference-LAPACK PR758)

Update the version number of the included LAPACK to 3.11.0

Improve convergence of LAPACK ?LAED4 and fix a bug in DORCSD2BY1 (Reference-LAPACK PRs 655+697)

Fix documentation of LAPACK functions ?TPRFB and IEEECK (Reference-LAPACK PRs 665+735)

Simplify ?SYSWAPR and fix its documentation (Reference-LAPACK 217)

Use new algorithms for computing Givens rotations (Reference-LAPACK PR631)