* move segment_pool to phi.

* mark summed ids as optional tensor.

* fix as reviews.

* move gather.h gather.cu.h scatter.h scatter.cu.h to phi library

* fix CI

* fix rocm ci

* 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 operators/math/math_function_* to pten/kernels/func
* namespace from `paddle::operators::math` to `pten::funcs`

From --ptxas-options=-v, SegmentOpsKernel uses 66 registers in a block.
There are two ways to resolve this problem:
    Reduce the threads per block launch configuration
    add __launch_bound__ to give information to nvcc compiler for reducing registers usage
this PR chooses __launch_bound__ solution because changing gpu_launch_config may affect other ops.

* refine structure for cuda and rocm

* update

* update

* update

* update

* [OPs] Bug fix, fix the segment mean for illegal syncthreads usage.

fix cpplint error for the autmic max/min

Add GPU Kernels of Segment Ops,  support, sum, max, min, mean