* Add yolov4 example option.
Add yolov4-tiny for benchmark.

* Restyled by astyle

* Restyled by clang-format

* Restyled by astyle

Co-authored-by: Restyled.io <commits@restyled.io>

* tf2sb

* wip

* wip

* concat maxpool avgpool

* StridedSlice

* handle stridedslice mask

* convert mlir deconvolution

* convert deconvolution output shape

* convert tanh

* code format

* Support yolov4-tiny conversion.
Add darknet2ncnn readme.
Add yolov4 example.

* Restyled by clang-format

* Restyled by astyle

* Restyled by clang-format

* Restyled by astyle

Co-authored-by: Restyled.io <commits@restyled.io>

* update roialign

* update tool for roialign

* fix ceil for roialign

* fix ios build

* fix

* fix

* make it compatibile with the old version

* fix ios build

* trigger CI

* fix test

* order

* build

* remove code

* merge roialign

* accelerate ROIAlign

* update note

* rename func

* roialign version

* trigger CI

* fix roialign

* use ref for pre-calc in roialign

* retrigger CI

* pre-alloc

* fuse mish and depthwise convolution

* darknet2ncnn: support enet-coco conversion.

* darknet2ncnn: Add an option, which is enabled by default, to merge all yolo output into one layer. The convert result is compatible with yolov3 example and can be used in benchmark.

* Add darknet to ncnn converter with YOLOv4 support.

* darknet2ncnn: add copyright info, fix clang compatibility

* fix wrong target

* darknet2ncnn: compatible with clang++ 9.0.0

Co-authored-by: zcy6735 <zcy6735@arcsoft.com.cn>

* lstm with direction

* convert onnx lstm

* lstm works

* use Mat class for Shape description

* shape specialization constant in compute shader

* wip

* wip

* test forward_inplace, add binaryop unaryop sigmoid test

* fix arm unaryop test

* fix arm binaryop test

* make shape hint optional, cast int8 to fp32, add cast test

* wip

* follow the good and old local size setting for conv1x1

* the optimal local size rewrite

* fix build on msvc

* add permute shader for all packing layout, add permute test

* concat and slice patial shape constant, slice test

* fix slice test

* interp test

* add lrn test, test packing layout implicitly

* add eltwise test

* add normalize test

* add instancenorm test

* reorg shape constant

* simple local group size partition

* add shape constant param

* fix type conversion warning

* fix MSVC _CRT_SECURE_NO_DEPRECATE warning

* type conversion: missing one

* fix  MSVC _CRT_SECURE_NO_DEPRECATE warning and type warning

* reformat

* convert google::protobuf::int64 to size_t to avoid type impl between msvc and gcc gap

* remove C++11 auto

* tab -> space

* revise more understandable help info

* try update ubuntu system before apt-get install

* remove upgrade