【doc and opt tool】add auto_transform.sh into Paddle-Lite project and modify related docs #2973 (#2974)

* 【arm】fix pooling no-equal padding problem (#2956)

* test=develop
Co-authored-by: NHappyAngel <chenjiaobuaa@126.com>

fix: move quant op to basic

* [Doc] Add post training quantization and weight quantization docs (#2960)

skip fusing quantized conv2d + relu6 for now

 fix xcode fcmp build error

add multiclass_nms2 kernel, test=develop

* [doc]add java api doc and python api doc (#2935)

* add docs for cxx demo and java demo, test=develop, test=document_fix (#2932)
* [cherry-pick] add all other docs for release2.3

* update support hardware md (#2930)

* fix: update fpga doc

* add weight quant for combined model after optimized, test=develop

test=develop

* add layout, opencl docs. test=document_fix

* fix link in doc. test=develop, test=document_fix

* mobilenetv1/v2,mnasnet,yolonano,sr models passed with opencl image2d kernel. test=develop

* Fix gflag shape define. test=develop

* comment All opencl buffer kernels. test=develop

* refactor conv, depthwise into one routing selection. test=develop

test=develop test=release/2.3

fix fpga lite_tensor compile bug
     add fake quantize_abs_max op
     test=develop

Update mask demo docs, add model download from paddlehub

* [LITE][BM] support VGG,Inception,Mobilenet,Darknet, test=develop

* modify opt name test=develop

* fix code style test=develop

* test=develop

* replace gru RowWiseAdd Eigen with c implementation
test=develop