Add canonicalization pass for torch import (#18150)
It is possible for post decomposition that there will be available
canonicalizations. Added the pass to support the cleanup.
diff --git a/compiler/plugins/input/Torch/InputConversion/Passes.cpp b/compiler/plugins/input/Torch/InputConversion/Passes.cpp
index 8f51c61..00ab1a4 100644
--- a/compiler/plugins/input/Torch/InputConversion/Passes.cpp
+++ b/compiler/plugins/input/Torch/InputConversion/Passes.cpp
@@ -50,6 +50,7 @@
pm.addNestedPass<func::FuncOp>(
torch::Torch::createDecomposeComplexOpsPass(emptyArrayRef));
pm.addNestedPass<func::FuncOp>(torch::Torch::createFuseQuantizedOpsPass());
+ pm.addNestedPass<func::FuncOp>(createCanonicalizerPass());
pm.addNestedPass<func::FuncOp>(torch::Torch::createScalarizeShapesPass());
pm.addNestedPass<func::FuncOp>(torch::createConvertTorchToTMTensorPass());
pm.addNestedPass<func::FuncOp>(
