* Convert CPU ops to new form

Remove obsolete sigmoid

* Fix export

* fix typo

* typo

[Py] Added clamp operator to Python API

* Convert PlaidML ops to new form

* style

* Add CF for Gather

* Style

* Move SlicePlan out of DynElimination, for reuse in ConstantFolding

* Add CF support for Slice

* Add CF for DynSlice

* Add <algorithm> to slice_plan.cpp, to make Windows happy

* Update Reshape CF to support all ETs

* Add CF for DynReshape

* Add CF for Transpose

* Add #include <numeric>, for std::iota

* style, oops

* Add Output support for a number of builders

* ../src/ngraph/builder/autobroadcast.cpp

* Remove some _values

* Simplify

* Simplify, add some more ops/builders

* ONNX failures

* Some GetOutputElement changes to help with Output<Node>

* Convert some ops to use Output<Node> inputs

* Review comments

* Some more changes of nodes to outputs

* Chane names for OutputVector helpers to avoid API change

* Cleanup

* Some more changes of nodes to outputs

* Chane names for OutputVector helpers to avoid API change

* [MLIR] Bump MLIR repo to commit 59167c2, 07/25/2019.

MLIR commit:
Author: River Riddle <riverriddle@google.com>
Date:   Wed Jul 24 16:41:11 2019 -0700

    NFC: Use ValueOfRange instead of T in Diagnostic::appendRange.

        For iterator_range, T is often the name of another iterator type
        and not the the value of the range.

LLVM commit:
Author: Marshall Clow <mclow.lists@gmail.com>
Date:   Thu Jul 25 03:26:05 2019 +0000

    Implement change #4 of P1466: Change weekday to accept both 0 and 7
    as Sunday. Add accessors 'c_encoding' and 'iso_encoding' to provide
    different interpretations of the weekday. Remove 'operator unsigned'

* style

* Move MLIR/LLVM repos a bit more forward

* CF for Sign, and extend element type capabilities for unary arithop CF

* CF for Ceiling and Floor

* Update CPU CF builders

* Update CPU CF builders

* CF for Not

* Add tests for new CPU CF folders

* Add tests for recently added CPU CF functors

* Add tests for non-CPU ceiling/floor CF

* Unit tests

* Add test for CPU folder

* CF for Sign, and extend element type capabilities for unary arithop CF

* CF for Ceiling and Floor

* Update CPU CF builders

* Update CPU CF builders

* Add tests for new CPU CF folders

* Add tests for recently added CPU CF functors

* Add tests for non-CPU ceiling/floor CF

* Unit tests

* Add tests for new CPU CF folders

* Add tests for recently added CPU CF functors

* CF for And and Or

* CF support for comparison ops

* Fix predicate for binary elementwise; add unit tests for non-arithmetic binops

* Update CPU CF builders

* Make validation a pass and add it after every pass by default

* cleanup

* update per review comments

* Switch plaid to new API for disabling  pass validation

* address review comment

* [MLIR] Enable affine dialect loop fusion

Enable affine dialect loop fusion in nGraph pipeline. It also adds an
opt flag to enable/disable it when ngraph-opt is in place. Fusion seems
to work for simple cases. It wasn't able to fuse dot + add, though, at
least in my test case. One example that worked:

Input:
  %6 = alloc() : memref<2500x2500xf32>
  affine.for %i3 = 0 to 2500 {
    affine.for %i4 = 0 to 2500 {
      %7 = load %arg0[%i3, %i4] : memref<2500x2500xf32>
      %8 = load %0[%i3, %i4] : memref<2500x2500xf32>
      %9 = addf %8, %7 : f32
      store %9, %6[%i3, %i4] : memref<2500x2500xf32>
    }
  }
  %10 = alloc() : memref<2500x2500xf32>
  affine.for %i5 = 0 to 2500 {
    affine.for %i6 = 0 to 2500 {
      %11 = load %arg2[%i5, %i6] : memref<2500x2500xf32>
      %12 = load %0[%i5, %i6] : memref<2500x2500xf32>
      %13 = addf %12, %11 : f32
      store %13, %10[%i5, %i6] : memref<2500x2500xf32>
    }
  }
  %14 = alloc() : memref<2500x2500xf32>
  affine.for %i7 = 0 to 2500 {
    affine.for %i8 = 0 to 2500 {
      %15 = load %10[%i7, %i8] : memref<2500x2500xf32>
      %16 = load %6[%i7, %i8] : memref<2500x2500xf32>
      %17 = addf %16, %15 : f32
      store %17, %14[%i7, %i8] : memref<2500x2500xf32>
    }
  }

Output:
  %8 = alloc() : memref<2500x2500xf32>
  affine.for %i3 = 0 to 2500 {
    affine.for %i4 = 0 to 2500 {
      %9 = load %arg2[%i3, %i4] : memref<2500x2500xf32>
      %10 = load %2[%i3, %i4] : memref<2500x2500xf32>
      %11 = addf %10, %9 : f32
      %12 = affine.apply #map2(%i3, %i4, %i3, %i4)
      %13 = affine.apply #map3(%i3, %i4, %i3, %i4)
      store %11, %0[%12, %13] : memref<1x1xf32>
      %14 = load %arg0[%i3, %i4] : memref<2500x2500xf32>
      %15 = load %2[%i3, %i4] : memref<2500x2500xf32>
      %16 = addf %15, %14 : f32
      %17 = affine.apply #map2(%i3, %i4, %i3, %i4)
      %18 = affine.apply #map3(%i3, %i4, %i3, %i4)
      store %16, %1[%17, %18] : memref<1x1xf32>
      %19 = affine.apply #map2(%i3, %i4, %i3, %i4)
      %20 = affine.apply #map3(%i3, %i4, %i3, %i4)
      %21 = load %0[%19, %20] : memref<1x1xf32>
      %22 = affine.apply #map2(%i3, %i4, %i3, %i4)
      %23 = affine.apply #map3(%i3, %i4, %i3, %i4)
      %24 = load %1[%22, %23] : memref<1x1xf32>
      %25 = addf %24, %21 : f32
      store %25, %8[%i3, %i4] : memref<2500x2500xf32>
    }
  }

* Rename MLIR_LLVM_OPTIONS to NGRAPH_MLIR_OPTIONS

Something like this works now:
NGRAPH_MLIR_OPTIONS="--enable-affine-loop-fusion=false"

* Disable loop fusion by default and fix typo

[MLIR] Fix bad merge on 2 MLIR changes

[MLIR] Re-compile sub-graph once on first invocation

Changes to get allow NNP to pass tests

* CF for Sign, and extend element type capabilities for unary arithop CF

* Update CPU CF builders

* CF for Sum

* style

* CF for Concat

* Switch from Nodes to Inputs/Outputs

API is synchronous per thread and threads are coordinated so that
we know when we hit the last iteration everything is done.
Using join() to gate end of iterations was introducing too much
overhead to timing as verified via checking traces.