* wip

* using namespace cleanup

* wip

* wip

* remove get_vector from runtime::TensorView class as it was for unit test only

* cleanup

* move writting vector to runtime::TensorView to the unit test dir

* merge fix

* PR review change

* update from PR comment

* update changes file

* Function can have multiple results
Remove external use of ValueType, TupleType, Tuple
Remove many external uses of Output and Input

* corresponding CPU backend changes

* Update master changes.

* Remove type arg from Function, add changes.md

* Merge changes.

* Move bodies to .cpp, add brief doc

* Merge CPU changes.

* Remove xla includes from non-xla files

* Remove xla from tests

* First part of xla tuple support

* change fprop_cache to assume multi-output bprop functions

* New wrappers for handling tuples with XLA

* Review comments

* remove old xla files

* fix merge errors

* hand edit models to use multi output instead of tuples

* remove ngvm

* remove NGVM from cmake

* remove ParameterizedConstant
* use simpler element Type definition
* Move TraitedType to NGVM directory

* Add numpy_transpose to the builder for numpy-stype transpose operations

* fix docstring

* make sure to throw the error

* First compiling pass, doesn't link tests, looks like templates aren't properly initialized?

* Make reduce inline, compiles, sum passes test

* Initial working reduce op builder

* Add Std dev, Variance, L2Norm. Tests fail because Power isn't actually implemented in ngraph++ yet

* Add Doc Strings, slight refactor

* fix tests after master merge

* Fix style issues raised in PR review

* Switch to Sum-based ops instead of Reduce-based ops for simpler autodiff.

* Add requested TODO comments

* Use all_close for tests

* Remove product op