add concat op with CPU kernel

* init add

* add topk op

* someupdate

* fix style check

* add test py file

* update top k cuda kernel

* follow comments

* remove debug print

* fix casting error

* fix casting error

* fix casting error

* fix rename bug...

* fix travis

of identity to Y.

1. finish lookup table CPU and GPU kernel
2. Add some cuda helper
3. Add some math funtor

* 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

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.

Currently use `Operator("fc", X="x", W='w1', B='b1')` as operator
creation method.

Fix #3198

Add get_numeric_gradient API and its unittest.