As the title.

* replace with pten kernel in cast cuda compute and remove unused codes

* rm unused header file

* replace CastCUDAOpKernel with CastOpKernel

* Rearranged Eager AutoCodeGen directory structure

* Removed USE_OP in Eager AutoCodeGen

* Enabled generation for Operators without Grad/Inputs/Outputs

* Resolved operators without input

* Fixed merge conflicts

* Enabled Eager AutoCodeGen for 10+ more operators

* Refactored Eager AutoCodeGen with more organized helper objects

* Enabled Eager AutoCodeGen for operators with multiple OpBases

* Adjusted Eager AutoCodeGen to Enable Passing Output Tensor as Input Argument

* Handled Dispensable Inputs/Outputs in Eager AutoCodeGen

* Adjusted function generation/call between Python-C API & Dygraph API

* Synchronized auto-generated Python-C API with Dygraph Forward Functions

* Added safe_initialized interface to EagerTensor for use in processing dispensable inputs

* Enhance error msg in paddle.assign in static mode

* fix unittest

* add map_matmul passes to quant2_int8_mkldnn_pass

* fix fc+act fuse (activation scale)

* ci fix, c++17 structured bindings not available

* fix ci static check

* fix bug of set_value op

* fix BumpInplaceVersion

* polish some comments

* revert change of full_like

* update Planner

* update unitest

* update PlanSpace

* update PlanSpace

* modify set_grad_var_shape

* update code style

* polish register marco

* resolve compile failed

* revert needless change

* revert eager related change

* revert eager related change

* change register marco name

* polish deetails

* add conv_gelu_mkldnn_fuse_pass

* add post ops

* layer.to api support numpy.dtype and paddle.dtype

* skip the layer to eaxmple code executing for env not support

* Reduce reshape kernel functions in pten

* delete notes

* fix bugs when compile

* test_mkldnn_depthwise_conv_pass

* test: add TimeOut

* sset TIMEOUT

* fix:add random num for dilation and group

* Rearranged Eager AutoCodeGen directory structure

* Removed USE_OP in Eager AutoCodeGen

* Enabled generation for Operators without Grad/Inputs/Outputs

* Resolved operators without input

* Fixed merge conflicts

* Enabled Eager AutoCodeGen for 10+ more operators

* Refactored Eager AutoCodeGen with more organized helper objects

* Enabled Eager AutoCodeGen for operators with multiple OpBases

* Adjusted Eager AutoCodeGen to Enable Passing Output Tensor as Input Argument

* Handled Dispensable Inputs/Outputs in Eager AutoCodeGen

* add layer_norm_fuse_pass test case

* restore cmakelist code

* Merge branch 'develop' into layer_norm_fuse_pass

* Merge branch 'develop' into layer_norm_fuse_pass

* add bad case test

* add white list for dist passes

* update comment

* follow zhiqiu's comment

* fix PassContext attrs type

* reshape+transpose+matmul_v2

* in_name->input_name

* fix pr-ci-static-check

* make some test higher paralle level

* modify slightly acoodring ci test:40100-40107

* modify slightly acoodring ci test:40124-40134

* modify slightly acoodring ci test:40138-40150

* modify slightly acoodring ci test:40153-40165

* modify slightly acoodring ci test:40153-40165

* add_conv_bn_fuse_pass_test

* add trt NHWC bug into invalid program for avoiding error crash

* add trt NHWC bug into invalid program for avoiding error crash

* add trt NHWC bug into invalid program for avoiding error crash

* fix trt pass target optype bug based on has_bias value

* fix trt pass target optype bug based on has_bias value

* update 3 tests

* fix typo error

* disabled bad case

* int to size_t

* fix single card 8 unittests in new executor

* fix

* fix