Improve FuseMultiUseElementwiseProducersPass

[IanWood1]

The helper function isHorizontalToGroup relies on getBackwardSlice which doesn't include operations defined above which are used in the body of an operation (vs as an operand) https://github.com/iree-org/iree/blob/114a1427810f3da0234f98c22f58390773b0489a/compiler/src/iree/compiler/DispatchCreation/FusionUtils.cpp#L104

To be conservative, isHorizontalToGroup returns false when there is a value defined above. This could be improved to track values defined above and add them to the slice. This would also require a corresponding change to moveOperandDefs here (so that values defined above are also moved): https://github.com/iree-org/iree/blob/114a1427810f3da0234f98c22f58390773b0489a/compiler/src/iree/compiler/DispatchCreation/FusionUtils.h#L37

This would allow for cases like the following to be fused:

#map = affine_map<(d0, d1) -> (d0, d1)>
module {
  util.func public @fuse_by_moving_consumer(%arg0: tensor<5x5xf32>, %arg1: tensor<5x5xf32>) -> (tensor<5x5xf32>, tensor<25xf32>) {
    %cst = arith.constant 1.000000e+00 : f32
    %cst_0 = arith.constant 2.000000e+00 : f32
    %0 = linalg.generic {indexing_maps = [#map, #map], iterator_types = ["parallel", "parallel"]} ins(%arg0 : tensor<5x5xf32>) outs(%arg1 : tensor<5x5xf32>) {
    ^bb0(%in: f32, %out: f32):
      %2 = arith.addf %in, %cst : f32
      linalg.yield %2 : f32
    } -> tensor<5x5xf32>
    %collapsed = tensor.collapse_shape %0 [[0, 1]] : tensor<5x5xf32> into tensor<25xf32>
    %1 = linalg.generic {indexing_maps = [#map, #map], iterator_types = ["parallel", "parallel"]} ins(%0 : tensor<5x5xf32>) outs(%arg1 : tensor<5x5xf32>) {
    ^bb0(%in: f32, %out: f32):
      // Note %cst_0 defined above
      %2 = arith.subf %in, %cst_0 : f32
      linalg.yield %2 : f32
    } -> tensor<5x5xf32>
    util.return %1, %collapsed : tensor<5x5xf32>, tensor<25xf32>
  }
}