* some ci tweaks

* workaround windows-2016 cache

For compatiblity for old compilers such as ndk-r16b.
`fabs()` keeps result stable.

ref: https://github.com/Tencent/ncnn/issues/2372

* recognize apple M1

* macos ios cpuset and thread affinity

* big little cpu topology for macos and ios

* silence affinity not supported error

* Add readme for keras2ncnn

* Add supported model variants

* Fix supported model variants

* Add extract without convert pack/fp1632

* SSE2: BatchNorm

* Fixed batch norm in AVX configuration

* Optimized register size switch

* Attempt to pass CI

* Attempt to pass CI

* Bias op

* Element wise ops

* Support packing on x86 by default

* Fixed macro range in bias

* Use aligned read for packed data

* Update testutil.h

* Update pooling_x86.cpp

* Support wasn SIMD

* Fix emscripten compiler flags

* fix build

* more ci fix

* concat x86 pack4

* flatten x86 pack4

* more x86 pack4

* ci pass

* fix

* enable sse2 mathfun

* enable --experimental-wasm-simd

Co-authored-by: nihui <shuizhuyuanluo@126.com>
Co-authored-by: nihuini <nihuini@tencent.com>

* fix compiling failure on cxx14

* fix compile failure when NCNN_PIXEL off

* do not export pixel api when NCNN_PIXEL off

* disable squeezenet test when NCNN_PIXEL off

* do not build ncnn2table and examples when NCNN_PIXEL off

* make cmakedefine header independent

* restore platform.h.in

Co-authored-by: nihui <shuizhuyuanluo@126.com>
Co-authored-by: nihuini <nihuini@tencent.com>

* add interp param align_corner

add check support_vulkan after create_pipeline for tests

* code style

Co-authored-by: yangfengmax <yangfengmax@didichuxing.com>

* fix vs2017 nmake warning C4267: 'return': conversion from 'size_t' to 'int'

* fix vs2017 nmake warning C4267: 'initializing': conversion from 'size_t' to 'int'

* drop bug_layout_binding_id_alias flag

* openmp on webasm works

* fix compile flags

* dynamic kmp runtime initialization

* clang simpleomp ci

* fix dispatch on unique cpu