* optimize memcpy perf for kunlun (#30291)

* optimize memcpy perf for kunlun

* remove useless unitest for kunlun mean

* minor

* fix bug that cann't find mkldnn(kunlun) (#30394)

faster build by reduce by-product, reduce linking library and fix compile warning of std=c++11 (#22164)

TemporaryAllocator is a singleton used for allocating memory for Cudnn. Since it is a singleton, we can delete it for better performance in memory.

We replace TemporaryAllocator by CUDADeviceContextAllocator and CUDADeviceContextAllocation, which uses stream callback to delete the memory allocated for the stream to avoid singleton.

Also added data_feed_proto to operator to fix CI in CPU compilation

* Revert "Revert "Add Event for TensorCopy" (#16022)"

This reverts commit e2da3a5b.

* use default stream
test=develop

* Revert "Add Event for TensorCopy (#15953)"

This reverts commit 7235fd66.
test=develop

* fix CI
test=develop

Add Event for TensorCopy 

* Revert "Add Event for TensorCopy (#15953)"

This reverts commit 7235fd66.
test=develop

* fix CI
test=develop

Add Event for TensorCopy 

test=develop

test=develop

Use OO style to rewrite memory allocation.

* make enforce a target and dependent on nccl when gpu is enabled

* add some more dependency

* Add debug logs in scope, meta_cache and memory

* Add missing deps