* - Added graph pass for fusing RNN op across layer
- Added test case for inter v/s cpu for verifying layer fused RNN
- more sanity checks in the RNN fusion graph pass
- added support to replace the recurrent cell state correctly in the fused RNN op

* Fixed multi layer rnn fusion unit test failure

* Addressed PR comments

* enable tests

* add funciton call

* working version

* remove test from ski list

* remove comments within body of function

* change to emit each op exactly once

* update code per review comments

* backend libraries now found in tree

dynamically read header search paths

fix running from install

*  group conv init

* add GroupConvolution op; refine checks in fusion logic

* add an emitter, cpu assigment

* cpu_layout

* add checks to algebraic simplification

* updating emitter logic for groupconvolution

* working before refactoring

* moving primitive creation logic to mkldnn_emitter

* group convolution graph test

* rename an opt

* address jbobba's feedback

* add pad_dilation function

* add dilation to gpu_emitter

* add CoordinateDiff constructor to GPUShape

* remove unecessary cast

* working version for forward

* forward working

* forward test all pass

* deconvolution forward

* backward data dilation

* forward test passed

* initial to 0

* fix bug for get_padded_shape and clang format

* code style, change variable names

* refactor convolution conditions

* fix bug padding_below_diff

* change pad_dilation to pad_dynamic, compare to pad

* remove passed convolution test from skip list, clang format

* change pad to use GPUShape

* Added op::ReplaceSlice and enabled respective tests.

* div64 -> division_by_invariant_multiplication

* Added GPUMemoryManager for aggregating memory allocations and copies into a single operation for kernel arguments, and a reusuable memory space for workspace allocations.

* Added GPUShape and reworked Shape helpers to be
compatible with different shape types.
Shape is now implicitly convertable to GPUShape.

* Updated shape helpers signature and add conversion operators/constructors for GPUShape.

* Removed several unecessary static_casts now that GPUShape is utilized. GPUTensorViewWrapper had a few functions returning std::vector<size_t> instead of Shape/Strides. These were updated as well to take advantage of GPUShape convertion operators.

* Forgot to fix lambda for workspace allocations to match that of argspace allocations.

* Added GPUShape and reworked Shape helpers to be
compatible with different shape types.
Shape is now implicitly convertable to GPUShape.

* Updated shape helpers signature and add conversion operators/constructors for GPUShape.

* Adjust row_major_strides to avoid reversed-copy.

* Moved declaration out of loop for clang.

* Moved gpu_shape to gpu transformer.

* Removed no longer necessary headers.

* Added stdexcept header to gpu_shape.hpp

* Coordinate->GPUShape

* Refactored replace_slice into CudaKernelBuilder. Simplified allocations using new GPUAllocator and GPUMemoryManager.

* Refactor allocations to make use of primitive emitter.
Now memory primitives are registered at compile time and
the gpu memory address is resolved at runtime by ivoking
the primitive.

* Changed check on 64bit shape to check if high bits are set.

* Added const qualifier to data being copied in GPUAllocator::reserve_argspace

* Added const qualifier to data being copied in GPUAllocator::reserve_argspace

* Replaced runtime host to device memcpys with GPUAllocator reservations in order to move them to compile time.

* Forgot to remove no longer necessary buffer freeing from op emitters.

* Removed replace slice.

* Removed more replace_slice diffs.

* Updated replace_slice op to utilize GPUShape and GPUMemoryManager.

* Added back missing changes after timeline resolution.

* Added spacing between functions in GPUShape and boolean operators in shape.hpp.

* Template parameters are UPPER_SNAKE_CASE.

* Added unit tests for GPUMemoryManager and added checks that ensure the
device memory is allocated prior to address resolution by the memory_primitives.
Also exposed the allocation size of the memory manager.

* Return type of shape_size should be large enough to encapsulate the full stride of the tensor.
This should be 64bits wide regardless of the underlying value_type of the ShapeType.

* Upstreaming changes to shape_size (which returns size_t).

* cuDNN softmax impl. for all axis activation.

* Added catch for per-axis activations.

* Removed commended headers.

* Added explicit function for queueing kernel argument data rather than inline in the reservation function per @fengleitian recommendation.

* Add softmax cuda kernel. It relies on atomic memory addition to global
memory, this will add contention and should be optimized in the
future. A multilevel reduction can be found in
cs/gpu_softmax_cuda_shfl but it requires some further engineering.

* Refactored reduce coordinate transform code into a helper and applied it to broadcast.
Broadcast added to CUDAEmitter, now supports multiple non-consecutive axes.

* Removed change to data_types variable and updated/removed comments.

* Refactored softmax into the emission of two fused elementwise collective ops.
Added fused elementwise + collective kernels. Softmax is then just the combination of exp_sum_reduce + div_broadcast.

* Added default param to GPUAllocator::reserve_workspace to request memory initialization for each invocation of the memory primitive.

* GPU workspace memory is zero initialized by default but can be turned off if desired.

* Added template parameter to CUDAEmitter::build_elementwise, REDUCE_OP_TYPE,
to specify the ngraph op type to use for the reduction in the fusted ew_collective kernel.

* Renamed variables and updated a comment.

* Removed outdated softmax kernel to avoid confusion. Can be added later when atomic reduce is replaced.

* Clang complained about lack of explicit destructor for AxisSet. Since cuda_emitter doesn't need AxisSet specifically, switch to std::set<size_t>.
This also has the benefit that in the future, if we wish to emit kernels without ngraph core (for example in a standalone binary via a
serialized graph manifest, we don't depend on AxisSet.

* softmax -> broadcast in build_broadcast.

* Separate elementwise and elementwise_collective.

* Added default param to GPUAllocator::reserve_workspace to request memory initialization for each invocation of the memory primitive.

* GPU workspace memory is zero initialized by default but can be turned off if desired.

* finally have something almost acceptable

* Optimized eigen kernel for 2D reduction on a 4D tensor used for spatial mean

* revert change to serializer

* CPU: Direct Execution
Part 1 with bare minimum infrastructure

* Refactor: Move build related functionality to a separate TU
and external function method

* Add TU back after merge

* Remove an assert

* Remove commented-out code

* batch dot pattern wip.

* batch dot pattern wip.

* added batch dot op.

* batch dot compute testing.

* correct gemm parameters.

* renaming matrix fusions passes and update tests.

* clean up.

* clang format.

* more clean ups.

* clang format.

* added CPUBatchDotFusion to default cpu passes.

* added missing header.

* added element type check.

Fix proxy settings in contrib/docker so that builds work both on the Internet and within Intel (#1088)

* Added conditional code to only set proxies when hostname appears to be on an Intel network (.intel.com)

* Replaced Intel-network-conditional proxy setting code with a system that checks for existence of http_proxy and https_proxy, like the Makefile does

* Applied fix for NGRAPH-1862, as I ran into NVidia dockerfile hangs.  Temporarily use --no-cache with docker-build, to check that proxy fix really works (and we don't get a cached version).

* Restored an original line I accidentally deleted.

* Remove --no-cache, which I had added for thorough testing.

* Support 3-D pool with mkldnn

* Move execute() to test_tools.hpp

* add StopGradient op

* add StopGradient op src

* remove adjoints and add interpreter

* fix compile issue

* use nop_elimination and add unit-test

* update cmake

* update unit-tests

* slice elimination

* add comment for simplify_concat

* fix concat_slice

* another reshape-related fix

* added a missing header

* disable reshape-concat optimization

* test fix

* strip off attributes from backend type prior to shared lib dlopen

* change convolution to use cudnn emitter

* convolution working

* add asymmetric pad

* forward with asymmetric working

* backward asymmetric

* padding to padding_below

* pad still has bug on backward

* change name

* fix convolution back prop

* fix code block

* slice back from padded output:

* working code

* extra ,

* Update gpu_emitter.cpp

* splict build_convolution to 3 function

* format and fix bugs

* Update cudnn_emitter.hpp

* Added per argument alignment to GPUAllocator::reserve_argspace.

* Changed alignment in tests to match update to alignment in backend.

These tests should fail on all systems but they are somehow passing on the CI system.

* Update cmake files to more modern approach

* disable building libraries that are not required

* handle more build cases

* add versions to backend libs. add start of package target.

* add create_backend to backends

* temporary workaround to tbb not linking correctly with gcc

* install codegen lib

* force tbb to link to the cpu backend so that it is available for codegen

* fix clang build error

* fix warning for codegen build

* update cuda header paths

* change error message for opening backend shared library

* set lib path

* update serializer for all new ops

* add reduce op

* hack solution to get reduction function in reduct op

* hack version working on all tests

* add recude_window op

* fixed the reduction checking process

* add reduce window op, save progress, not compilable yet

* change puchback to pre allocate for vector

* fixed datatype vector

* dataype and comments

* pass op intead of using template

* using new GPUshape and allocator

* using GPUShape

* add comment, change map inside function.

* change to more menaful name

* refactor cpworkspaceinsertion for mxnet

* rename mxnet functions to adhere to ngraph naming convention

* define a member static const int in a cpp file to resolve a linking issue