Fix bug of foward default attribute not passed to backward

Merge infershapecontext and ExecutionContext

Fix flushing vars

* Implementing the Adamax optimizer step operator
* Adding unit tests for adamax_op

* Changing learning rate and time step to inputs from attributes

* Changing learning rate and time step to input(tensors)

* Making the Adamax operator conform to naming convention

* Removing Tensor<float> from comments

* Rectifying the Adamax implementation

* Changing Unit Test values and adding comments

* Changing Unit Test to test multiple steps

Unify Reduce functions and simplify register code

Build mobile inference library for minimum size

Backward for op desc

* Update program.md
* Update
* Update

* fix block syntax

* add builder

* fix argument names

* uniform all Var to var

* fix c demo

* factor set_outputs

* update block

* delete as_ifelse_input interface from block design

* simplify set_outputs and output_num

* make the c++ codes compatible with demo

* fix compatible

* fix syntax

* update ie

* update syntax

Implementing tanhshrink operator

Update if_else_op.md 

Complete register gradient for compile time

Add compile time infershape

Correcting grammatical mistakes in the python_api doc

Add persistable in framework.proto

Implementing the Adagrad optimizer step operator

Adding the implementation for rmsprop operator