* Split op skeleton

* Two ways to construct a fused Split to be able to use it in onnx importer

* refactor: move the util::split() helper functions to the core

* Split's decompose_op() implementation using a helper function

* Use fused Split in the onnx_importer

* Code formatting

* PR feedback

* Split helpers moved to ngraph/builder

* Basic UT - split a 1D tensor to 3 equal parts

* UT: Split 2D tensor into variable length parts

* Code formatting

* Catch the proper type of exception in the onnx_importer split()

* Initialize members in the correct order

* Type prop tests for Split

* Code formatting

* PR feedback

* Create mkldnn primitives at first iteration for CODEGEN.

 OPs: add, lstm, and rnn.

*  OPs: batchnorm.

*  OPs: concat and lrn.

Remove dead code.

* Skip in place concat, relu, reshape, and slice when building node_primitive_string_deps_index map.

* Change NGRAPH_ASSERT to NGRAPH_CHECK.

* Address PR Feedback.

* Create mkldnn primitives at first iteration for CODEGEN.
 OPs: convertlayout, relu, leakyrelu, boundedrelu, sigmoid, softmax, slice.

* Fix bugs.

*  OPs: quantizedconcat.

Check if there are descriptors before emitting code to read desc_file.

*  OPs: convolution backward.

Use macro to write mkldnn memory dims to generated file.

*  OPs: MaxPoolWithIndices and MaxPoolWithIndicesBackprop.

Add unit tests for MaxPoolWithIndices, MaxPoolWithIndicesBackprop, and MaxPoolBackprop.

* Fix style error.

*  OPs: AvgPoolBackprop and MaxPoolBackprop.

Add unit test for AvgPoolBackprop.

*  OPs: DeconvolutionBias.

*  OPs: Quantize and Dequantize.

*  OPs: QuantizedDot and QuantizedDotBias.

* Use reference kernel for QuantizedConvolution for CODEGEN when mkldnn does not support the parameter types.
Get scales for quantization ops in cpu_emitter.

* Fix Windows build error: add CPU_BACKEND_API.

* Use template for quantization ops.

*  OPs: QuantizedMatmul.

Emit referece kernel for QuantizedDot in CODEGEN.

* Remove QuantizedDot from get_scale_index.

* Address PR feedback.

* Add move operations to AlignedBuffer

* unit test

* Remove parent from PlaidML tensor initializer

* Remove plaidml tensor parent plumbing

* style

* remove throw from dtor (#2854)

* add special handle for the 'NNP_xx' op name

* style

* add special handle for the 'NNP_xx' op name

* style

* use description as suggest by bob

* Add builders for QLinearMatmul for onnxruntime

* Generic Dot

* Change the onnx bridge and address PR feedback

* Fix date

* Fix CI failure

* Change varibale filter to input1

* const & reference

* update branch

* Comment

* Introduce QuantizedMatmul

* change pattern

* QDot tests passing

* style

* PR feedback

* Fix pattern

* style

* Don't use
pragma GCC diagnostic error "-Wswitch"
pragma GCC diagnostic error "-Wswitch-enum"
for GCC 4.8.X

* Fix typo.

* Style.

* SquaredDifference implementation

* Broadcast input before using it

* Simple test of SquaredDifference

* SquaredDifference validation tests

* Formatting adjustments

* Docs correction

* Exclude the unit test on iGPU

* Keep the includes in a single group

* Update intelgpu_backend.cpp

* Update unit_test.manifest

* UT for the broadcasting path

* Add test for i32 gather

* Add support for ints to Gather op

* Move helper function to anonymous namespace

* Add more types

* Use static_cast instead of the old one

* Style fix

* Skip tests on GPU

* Add more tests

* Skip tests on gpu

* Change bool to char

* Supports more cases for Convolution Bias fusion and also removes redundant namespace qualifiers in core_fusion.cpp

* added unit test

* select all fusions

* constexpr ctor for EnumMask

* added pass properties to core passes.

* added unit tests.

* minor fixes.

* Extend lp-norm functions to take bias.

* Move lp-norm utilities to nGraph core op/util.

* Move norm files to builder directory.

* Apply clang-format.

* Sceleton for GRN operation.

* Add GRN implementation.

* Fix reshape utility function.

* Address review comments.

* Add using namespace std.

* Review comments.

* Few fixes in grn implementation.

* Clang format.

* Basic UT.

* Fix expected data.

* Add more UT and skip them on IGPU.

* Review comments: const correctness and remove using namespace std statement.

* Unblock GRN on IGPU.

* Get back GRN op case to switch.

* merge error

IntelGPU backend: Make ScaleShift op as not supported in callback of FusedOpDecomposition pass (#2940)

* IntelGPU backend: Make ScaleShift op as not supported in callback of FusedOpDecomposition pass

* style

* merge error

* Style was correct.  Return back 0965fb5 changes

* merge error

This PR removes compiler warnings like this one:

```
src/ngraph/frontend/onnx_import/utils/reduction.hpp:73:28: warning: prior to the resolution
  of a defect report against ISO C++11, local variable 'op_node' would have been copied
  despite being returned by name, due to its not matching the function return type
  ('shared_ptr<ngraph::Node>' vs 'shared_ptr<ngraph::op::ArgMin>') [-Wreturn-std-move-in-c++11]
                  return op_node;
                         ^~~~~~~
```

IntelGPU backend: Change clDNN mode of Relu op to allow ConvolutionBiasAdd fusing inside clDNN (#2939)

* Add placeholder for doc versioning

* Finalize module to include tags of verified releases only

* Add fused Squeeze op

* Use fused Squeeze op in ONNX importer

* Update serializer

* Add unit tests

* Add type prop tests

* Change Squeeze signature to accept a dynamic input for axes

* Update src/ngraph/op/fused/squeeze.cpp
Co-Authored-By: Adam Rogowiec <adam.rogowiec@intel.com>

* Code review comments

* Fix failing unit test

* Code review comments

* Code review comments

* style

* Add op to iGPU backend

* Add op to iGPU backend

* Fix clang compiler warnings

* Remove unintended file.

* style

* Not part of PR

* Another extra closure ref

* More warnings from merges

* Lambda arg was used

* Remove remaining osx compiler warnings

* style

* Try to avoid compiler warning

* Same for the other test

* IntelGPU backend: Enable FusedOpDecomposition pass

* PR2912. Adress comments.