* remove cudaDeviceContext

* remove more template

* fix rocm compile

* unify cpu context

* fix init()

* delete test_device_context

* fix test_scalar

* rename pten_utils to phi_utils

* rename pten_utils target

* rename Pten to Phi

* replace pten with phi

* resolve conflict

* rename pten dir to phi

* rename namespace to phi

* rename infrt pten dir to phi

* resolve conflict

* rename pten to phi in cmake

* revert all infrt change

* change needed files

* fix infrt failed

* fix inference failed

* move bernoulli kernel to pten

* follow comments

* Migrate Dim and DDim from paddle::framework into pten namespace

* fix paddle::framework::Array

* fix framework::Array

* add cpu_context.

* update

* update

* update

* update

* update

* fix ci problem

* fix npu ci problem

* update

* fix ci compile

* fix PADDLE_ENFORCE and refine the description
test=develop

There are mainly following fixes:

- take `DeviceContext` as the template parameter of math functors and OpKernel instead of `Place`
- remove `eigen_device` interface in base class  `DeviceContext`
- remove `GetEigenDevice` interface in `ExecutionContext` and base class `DeviceContext`
- remove unused `platform::EigenDeviceConverter`
- rename `REGISTER_OP_GPU_KERNEL` to `REGISTER_OP_CUDA_KERNEL`
- rename `USE_GPU_ONLY_OP` to `USE_CUDA_ONLY_OP`

CompareOp can run on CPU even other operators are running on GPU, since
opeatations like comparing control flags should be performed only on CPU

Using DeviceContext, not Place to get stream

Fix TravisCI

* with unit-tests
* Also complete `memcpy`