[Transform][Fusion] yield fused producer if necessary

[Yun-Fly]

Automatically yield fused producer if it has multiple uses. E,g,

func.func @yield_fused_producer(%arg0: tensor<16x32x32xf32>) -> (tensor<16x32x32xf32>, tensor<16x32xf32>) {
    %cst_0 = arith.constant dense<2.000000e+00> : tensor<16x32x32xf32>
    %dest0 = tensor.empty() : tensor<16x32x32xf32>
    %0 = linalg.powf ins(%arg0, %cst_0 : tensor<16x32x32xf32>, tensor<16x32x32xf32>) outs(%dest0 : tensor<16x32x32xf32>) -> tensor<16x32x32xf32>
    %dest1 = tensor.empty() : tensor<16x32xf32>
    %1 = linalg.reduce { arith.addf } ins(%0 : tensor<16x32x32xf32>) outs(%dest1 : tensor<16x32xf32>) dimensions = [2]
    return %0, %1 : tensor<16x32x32xf32>, tensor<16x32xf32>
  }

When fusing powf into tiled consumer reduce, %0 should also be yield because it has two users. Otherwise, there would exist repeat powf computation.

[Yun-Fly]

This patch is also helpful to debug intermediate result when correctness check failed.

[Menooker]

why do we need this cast?

[Yun-Fly]

notifyOperationErased is just the member function of RewriterBase::Listener, which is the sub-class of OpBuilder::Listener.

[Menooker]

Offline discussed.