Merge pull request !5783 from DeshiChen/graph_kernel_1.0

Merge pull request !5841 from yao_yf/EmbeddingLookUpSplitMode_move_to_parallel

Merge pull request !5967 from lz/master

Merge pull request !5754 from Peilin/add-type-support-for-greater

Merge pull request !5858 from zhengjun10/master

Merge pull request !5959 from ling/bug

Merge pull request !5944 from vlne-v1/bug-cell-copy

Merge pull request !5948 from baihuawei/sample-seed

1. Update akg submodule
2. Refactor akg_kernel_build, akg_ascend_kernel_build, akg_gpu_kernel_build
3. Add akg_kernel_json_decoder to support converting kernel_json to AnfNode.
4. Add GraphKernel Cost Model. (mindspore/_extends/graph_kernel)
5. Add some GraphKernel passes to GpuSession, move these passes to backend/optimizer/graph_kernel.
6. Add global id for ir files.
7. Fix bug in ConstInputToAttr.

Merge pull request !5911 from wangshaocong/awareTraining_ci

Merge pull request !5936 from zhaoting/mobilenet

Merge pull request !5935 from guozhijian/update_platform_info_master

Merge pull request !5933 from ZPaC/master-fix-error-when-pserver-finish-training

Merge pull request !5903 from lvchangquan/master

Merge pull request !5901 from panfengfeng/fix_network_issue

Merge pull request !5938 from EricZ/fix_zoo_api

Merge pull request !5924 from caojian05/ms_master_vgg16_support_imagenet_on_ascend

Merge pull request !5881 from yepei6/mobilenetv2

Merge pull request !5913 from limingqi107/master

Merge pull request !5885 from vlne-v1/add-doc-for-composite-op

Merge pull request !5921 from zyli2020/refine_cmake_find_cuda_cudnn

Merge pull request !5866 from liangzelang/fix_global_step_error

Merge pull request !5806 from xiefangqi/md_delete_setdatasetsize

Merge pull request !5736 from fary86/add_device_specific_config_check

Merge pull request !5749 from amongo/SupportPyantiveMultiGrad

Merge pull request !5893 from fary86/fix_adam_lamb_bug

Merge pull request !5902 from VectorSL/format

Merge pull request !5900 from 徐永飞/master