* Fix timer of arm cpu profiler. test=develop

* Fix un-added op in cmake.test=develop

* fix cmake error

* fix cmake error, test=develop

* Fix pass sequence. test=develop

* replace option with lite_option. test=develop

* disable profile mode by default. test=develop

* Fix error option name. test=develop

* add stack op and add reduce_mean op and their unit tests

* modify stack op output name and modify the for loop in reduce_mean op

* add HasAttr for slice op

add nearest_interp cuda kernel for Paddle-Lite

* [NPU] add NPU supporting for Java API
test=develop

* [NPU] refine build script for NPU compiling
test=develop

* [NPU] fix compiling script for NPU
test=develop

test=develop

* fix ios build script, test=develop

* add ios tiny publish target, test=develop

* fix ios build scrip, test=develop

* merge build_ios.sh to build.sh, test=develop

* ios support BUILD_EXTRA, test=develop

* Add precision and persistable attrs for the tensor. And fix cxx light and full api demo.

* update precision2string methods. test=develop

* move the save logic to the front of the run in mobilenetv1_full_api.cc, test=develop.

* add comments for UpdateVarsOfProgram. test=develop

* add yolo_box_compute cuda

* move multiclass_nms(arm) to host

* add lod in scale op

* add yolo_box_cuda cmake config

* modify shuffle_channel_fuse and transpose_softmax_transpose_fuse to support run ssd model. test=develop

* reshape and transpose op don't have xshape output.

* modify yolo_box_compute_cuda, use tensor to manage cuda memory test=develop

* add yolo_box use kernel test=develop

* refine toolchain test=develop

* fix wrap compilation error

* fix yolov3 armv8 compilation test=develop

* revert to armv7 as default test=develop

* fix fpga compilation test=develop

* paddle lite cuda init
can run model with leaky_relu

* add the missing file.
test=develop

* add the load from memory interface.
test=develop

* refine this pr. fix comments
fix ci error
test=develop

test=develop

ad ops for faster rcnn, including affine_channel, anchor_generator, generate_proposals and roi_align (#1895)

* add ops for faster rcnn

* disable test for generate_proposals and roi_align, test=develop

* remove .swp file

* remove log in tensor slice

* finish the unit test for roi_align, test=develop

* add npu script and tester

* fix npu armv7 so and refine tests

test=develop

* update fix and refine log

test=develop

* refine npu generate api

* refine npu subgraph

* refine npu gen and clean code

* fix model laod

* refine node2rm in subgraph

* refine the build npu functions

test=develop

* modify cast op, test=develop

* modify cast op and remove warning in argmax_test, test=develop

test=developt branch

* add transpose-softmax-transpose fuse pass

test=develop

* enable supported lite-npu ops

test=develop

add x86 math:sequence_scale,sequence_padding,sequence2batch,sequence_pooling. test=develop (#1884)

* [NPU] fix conv2d npu bridge, supports bias from input map
test=develop

* [NPU] support more dimensions for the bias of conv2d NPU bridge
test=develop

2.add flatten2 for gpu.
3.add concat 4 inputs size for gpu.
4.fix pool.
5.fix transpose2
test=develop

* paddle lite cuda init
can run model with leaky_relu

* add the missing file.
test=develop

* [test=develop]1.fix crash when gpu op scale&elementwise_add input dim size equal 2
2.add gpu op mul

* fix code style test=develop

* add npu script and tester

* fix npu armv7 so and refine tests

test=develop

* update fix and refine log

test=develop