* skip_layernorm_op done

* add unittest

* slice op convertor support trt < 6

* skip_layernorm only work in ernie

* add fusion group test for backward and refine code

* add support for expression type convert and add cast Op support in fusion group

* Implement a common python unittest to test the ir passes.
test=develop

* Save the results in np.array and support to startup on CPU.
test=develop

* Fix the unittest.
test=develop

* Add check_program to check whether the optimized program is different from the origin one.
test=develop

* Remove the inferface all_ops.
test=develop

* Add exception test in pass_test.
test=develop