* decouple memory copy

* fix ci bugs

* fix ci compile bugs

* fix rocm compile

* fix ci bugs

* decouple_memory

* perfect memory utils

* fix ci bugs

* fix inference bugs

* fix custom test bugs

* fix converage bugs

* modify code according comment

* modify namespace

* deal with compile bugs

* add pinv check

* add unitest

* update unitest

* roll back

* fix not call stupid bug

* use context

* refine cmake of framework

* add deps for dense tensor

* fix deps

* remove alloc(ctx)

* add depends on mkldnn

* svd cpu forward

* svd gpu forward

* transfer the backward of svd

* remove cusolver in svd_grad

* svd kernel bug fix

* fix bugs

* fix bugs.

* fix bug