* make fast through to linear

* make fast through to linear

* add to do for later upgrades

* support build once for now

* Refine eager run_program OP for dy2st UT

* append run_program error string and refine run_program_grad

* remove some comments

* refine ConstructXGradTensors

* refine partial_program

* fix code for test_mnist.py train

* support quantify UT

* make __fake_vars and _double_grads to lazy

* fix comments

* adapt run_program OP for eager

* fix program_id

* refine code

* fix test