* Fixed gelu save_graph error

Added SaveTmGeluOp()

* Added gelu timvx

* Added layernorm operator

* Added layernorm timvx

* try to fix ci

* apply code-format changes

* Update build-and-test.yml
Co-authored-by: NBUG1989 <BUG1989@users.noreply.github.com>

* std::max_element in range [first, last)

* Comment out misleading output

* fix yolov4-tiny anchors bug

* fix timvx bug with deconv_depthwise

* Update timvx_executor.hpp

* Update timvx_clip.cc

* add op spatialtransformer l2normalization

* format code
Co-authored-by: Ncmeng <cmeng@openailab.com>

* Update crop timvx

Co-authored-by: Nqwang <qwang02@openailab.com>

* add pad + reduction timvx op

* apply code-format changes
Co-authored-by: Nqwang <qwang02@openailab.com>
Co-authored-by: Nclovking <clovking@users.noreply.github.com>

* Create code-format.yml

* trigger on push only

* Update code-format.yml

* apply code-format changes

Co-authored-by: Nqwang <qwang02@openailab.com>
Co-authored-by: NBUG1989 <248857878@qq.com>

Co-authored-by: NYour Name <you@example.com>

* update trt relu6/clip

* del wrong shuffle

* fix timvx int8 tensor

* Update timvx_executor.cc

* Update timvx_limit.hpp

* Update timvx_executor.hpp

* Update timvx_executor.cc

* Update tm_hrnet.cpp

* update mish/eltwise_exp
* update timvx dump_tensor

* add yolov4-tiny timvx example

* merge from master

* fix slice to shufflenet_v2

* change new line symbol to \n only

* init to version 1.4

* add test back

* fix default linking libraries relation

* tab to space

* adapt to new api header name

* fix armv7 fp16 building issue

* fix test issues

* fix building issue

* fix test building issue

* disable yolact

* add permission for execution

* update tim-vx to 1.4

* update workflow to 1.4

* update directories

* fix x86 tim-vx running crash issue

* Update TIM-VX plugin

fix reshape
add transpose/permute

* Update tim-vx plugin

fix reshape
add permute/transpose

* Add files via upload

* Add files via upload

* Add files via upload

* enable prelu op for TIM-VX backend

* refactor hardswish op