1. Added bfloat16 based dot as new API: shdot
2. Implemented generic kernel and cooperlake-specific (AVX512-BF16) kernel for shdot
3. Added 4 conversion APIs for bfloat16 data type <=> single/double: shstobf16 shdtobf16 sbf16tos dbf16tod
     shstobf16 -- convert single float array to bfloat16 array
     shdtobf16 -- convert double float array to bfloat16 array
     sbf16tos  -- convert bfloat16 array to single float array
     dbf16tod  -- convert bfloat16 array to double float array
4. Implemented generic kernels for all 4 conversion APIs, and cooperlake-specific kernel for shstobf16 and shdtobf16
5. Update level1 thread facilitate functions and macros to support multi-threading for these new APIs
6. Fix Cooperlake platform detection/specify issue when under dynamic-arch building
7. Change the typedef of bfloat16 from unsigned short to more strict uint16_t

Signed-off-by: Chen, Guobing <guobing.chen@intel.com>

* make building the bfloat16 BLAS functions conditional on BUILD_HALF

* pass the BUILD_HALF option to gensymbol

* Pass BUILD_HALF as a compiler define for dynamic_arch builds

There is no difference in the way libnames are handle between FreeBSD
and linux or sunos. FreeBSD and DragonFly prefers having sonames as well

From keno's PR #2350 - this avoids the potential hang in blas_thread_shutdown where we may wait for threads to exit while they are waiting on the loader lock from DllMain

* Update make.inc entries for LAPACK 3.9.0

Reference-LAPACK PR 347 changed some variable names and relative paths

* Update LAPACK to 3.9.0

* Add new functions from LAPACK 3.9.0

* Add new functions from LAPACK 3.9.0

* Restore LOADER command 

as it makes it easier to specify pthread as needed

* Restore LOADER

* Restore EIG/LIN prefixes in cmdbase

* add binary path to lapack_testing.py call

* Restore OpenMP version check

* Restore OpenMP version check

* Restore fix for out-of-bounds array accesses

from #2096

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

pool cleanup upon THREAD_DETACH which is needed when compiled with
USE_TLS=1.

icc has known problems with mixed-language builds that ifort can handle just fine. Fixes #1956

Conflicts:
	exports/Makefile

Its exclusion was an oversight on my part.

With this change, OpenBLAS builds and all tests pass on OpenBSD 6.2
using Clang. Tested on x86-64 only, with and without DYNAMIC_ARCH=1.

Rewrite NOFORTRAN conditionals

... so that they do not trigger accidentally when NOFORTRAN is empty/unset

Remove default FEXTRALIBS in NOFORTRAN case

* The DLL is named cygopenblas.dll, not libopenblas.dll
* The import lib (still called libopenblas.dll.a) is installed

Android does not support versioned SONAME entries, ref. #1173

If you want to build LAPACK deprecated functions since LAPACK 3.6.0

make BUILD_LAPACK_DEPRECATED=1

illumos fixes

use different names for `openblas*.renamed` between osx and other unices, fixes
```
Makefile:121: warning: overriding commands for target `../libopenblas64_p-r0.2.15.a.renamed'
Makefile:100: warning: ignoring old commands for target `../libopenblas64_p-r0.2.15.a.renamed'
```

also clean `*.renamed`

2) If OpenBLAS cannot find Fortran compiler, it will only build BLAS
(without LAPACK).

for adding a prefix or suffix to all exported symbol names in the shared library
Useful to avoid conflicts with other BLAS libraries, especially when using
64 bit integer interfaces in OpenBLAS

Note that since OSX does not have the objcopy utility, setting these options
to non-empty values on Mac requires the objconv tool, available (GPL license)
from http://www.agner.org/optimize/#objconv