Signed-off-by: zhoufeng <zhoufeng54@huawei.com>

Merge pull request !2720 from xychow/fix-assign-in-while

Merge pull request !2775 from ZiruiWu/better_get_dataset_size_in_generator

Merge pull request !2772 from Jamie/removalpass

cpplint

Merge pull request !2602 from JesseKLee/cache_op_stage2

Merge pull request !2766 from yanghaitao/yht_serialize_num_parallel_worker

Merge pull request !2760 from JesseKLee/gcc9

Merge pull request !2762 from YuJianfeng/master

Merge pull request !2746 from lichen/remove_32byte_limit_for_cpu_gatherv2

Merge pull request !2699 from zhaozhenlong/op/scatter-max-vm

add copyright

Merge pull request !2756 from wenchunjiang/fix_control_sink_bug

Merge pull request !1904 from riemann_penn/add_indexed_slices

Merge pull request !2753 from xiefangqi/xfq_fix_docstring

Merge pull request !2743 from ougongchang/summary_collector_ut

Merge pull request !2741 from jiangjinsheng/issue_fix4

Merge pull request !2757 from guozhijian/update_release_notes

Merge pull request !2752 from kisnwang/fix-mix-session-device-id

Has fixed collecting optimizer error when mode is eval

Merge pull request !2732 from gengdongjie/master

Merge pull request !2744 from liuwenhao/inplace

Merge pull request !2748 from ghzl/change-order-params-only-equal-to-group-param

Merge pull request !2742 from zhaoting/master

Merge pull request !2745 from wangqiuliang/fix-invertpermutation-operator

Merge pull request !2701 from wangnan39/distinguish_cpu_sparse_operators_from_tbe

Merge pull request !2715 from zhoufeng/empty-graph-not-process

Merge pull request !2726 from yangyongjie/master

Merge pull request !2730 from hewei/tensor_lazy_allocation

1. Tensor data is lazy allocated if no data to be copied;
2. Add prefix to Tensor id to distinguish with other id.

Merge pull request !2677 from laiyongqiang/eos_commit

Merge pull request !2697 from kisnwang/format-device-ascend-code