* refactor register mechanism, current so size: 1.20MB. test=develop

* fix KernelRegistry::Global().Create. test=develop

* fix cpplint errors. test=develop

* fix test_subgraph_pass bug. test=develop

* register kernel with target,precision,datalayout combination. test=develop

* fix test_paddle_api no op found bug. test=develop

* enhance comment

* fix lite/kernels/arm/elementwise_compute_test.cc. test=develop

* fix code style

* revert format of unchanged files. test=develop

* fix code format according to cpplint 1.5.1. test=develop

* remove redundant include header. test=develop

* add op affine_grid. test=develop

* fix concatt axis < 0 errorr,ttest=develop

* fix format. test=develop

* Add retinanet_detection_output op, test=develop

* [ARM] add where_index op, test=develop

* [ARM] add where_index op, test=develop

* fix: winograd support unsame pad
test=develop

* feat: add winograd int8 kernel
test=develop

* fix: style fix
test=develo

* fix winograd_int8 ut sgement default. test=develop

* close basic_test, test=develop
Co-authored-by: NMyPandaShaoxiang <txg4794@163.com>

* add grouup_norm

* fix format. test=develop

* fix xiaodu crash. test=develop

* add deformable Conv op 

* fix ut, test=develop

* fix format. test=develop

* test=develop

* delete unuseful message. test=develop

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

* [OPENCL] develop expend opencl kernel & unit test ,test=develop

* [OPENCL] develop expend opencl kernel & unit test ,test=develop

[LITE][OPENCL] Add local work size, global work size for opencl profiler; fix bug for variable inference pass. test=develop (#3741)

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

int8 direct_conv, dw_conv add relu6 and leaky relu fusion

* improve 3x3s1 direct profile

* add gemv+relu6/lleakyRelu

* fix relu6 problem, test=develop

* fix format, test=develop

* add six / scale , test=develop

* reduce .so size. test=develop

* compile all targets when LITE_ON_TINY_PUBLISH=OFF

* unordered_map is more convenient when key is customized class

* test=develop

* [LITE][DOC] Add kernel profiler doc for add operation. test=develop

* add argmax profiler info. test=develop

* fix pooling bug and speed

* add 2x2s2p1 pooling. test=develop

* fix conflict, test=develop

* [LITE][PROFILE] Enhance ARM CPU profiler with real backend kernel. test=develop

[LITE][OPENCL] Enhance Profiler for OpenCL with in/out/filter shape, macs/macs_ps, real backend kernel etc. (#3641)

* [LITE][OPENCL] Enhance Precision Profiler for OpenCL. test=develop

movs unsqueeze frm arm to host.

[LITE][XPU] 1. Add precision switch(int16/int31) in XPUMultiEncoderOp; 2. Fix identity_dropout_eliminate_pass, |AttrType| of 'is_test' in OpDesc can be INT or BOOLEAN; 3. Enhance |__xpu__multi_encoder_fuse_pass|; (#3596)

* [LITE][XPU] Add precision switch(int16/int31) in XPUMultiEncoderOp

* [LITE][XPU] fix identity_dropout_eliminate_pass, |AttrType| of 'is_test' in OpDesc can be INT or BOOLEAN

* test=develop

* [LITE][XPU] suppress linkage error
test=develop

* [LITE][XPU] 1. Reorder |identity_dropout_eliminate_pass| before |__xpu__multi_encoder_fuse_pass|; 2. Enhance |__xpu__multi_encoder_fuse_pass|, it works well in more scenarios;
test=develop

* [LITE][XPU] Remove XPUConfig
test=develop