Fix #3582

* Also simplify pybind implementation by using OperatorBase as holder
  type.

* make stepnet member

* add pybind support

* fix Inputs Outputs

* remove unique_ptr

* Give which parameter, which element are wrong. And what max_diff is.

* init grad op checker

* can run

* add GradeChecker class

* use get_numeric_gradient

* refine code

* add softmax and cross entropy auto grad test

* use close to judge op_grad and numeric_grad

* add cpu and gpu compare

* add comments

* add support_gpu

* fix allclose

* fix name error and symplify code

* optimize gradient checker

* add test_cross_entropy_op

* update gradient_checker.py

* optimize code

* use random.uniform instead of random.random

* fix type bug

* optimize check_grad

* put SupportGPU into OperatorBase

* typo

* Although backward_test/rnn_test is not pass, just comment them.

* check INFINITY in cross_entropy

* fix error

* use onehot_cross_entropy without GPU kernel

* add support_gpu

* fix allclose

* fix name error and symplify code

* add support_gpu

* fix allclose

* fix name error and symplify code

It can be run both CPU/GPU. configure attributes are:

* min: the min value of uniform random
* max: the max value of uniform random
* dims: the dimension of output tensor
* seed: the random seed of uniform random. 0 means generate a seed each
        time.

* move net_op to operators

Fix #3171

Add get_numeric_gradient API and its unittest.

* Make interface of Operator to `const Scope&`