fix(//core/lowering): Conv2D -> _convolution pass was triggering conv

transpose instead of conv

Signed-off-by: Naren Dasan <naren@narendasan.com>
Signed-off-by: Naren Dasan <narens@nvidia.com>

Author: narendasan

core/lowering/passes/conv2d_to_convolution.cpp

@@ -14,10 +14,9 @@ void Conv2DToConvolution(std::shared_ptr<torch::jit::Graph>& graph) {
             return (%4))IR";
     std::string convolution_pattern = R"IR(
         graph(%x, %w, %b, %s, %p, %d, %g):
-            %1 : bool = prim::Constant[value=1]()
+            %1 : bool = prim::Constant[value=0]()
             %2 : int[] = prim::Constant[value=[0, 0]]()
-            %3 : bool = prim::Constant[value=0]()
-            %4 : Tensor = aten::_convolution(%x, %w, %b, %s, %p, %d, %1, %2, %g, %1, %1, %3)
+            %4 : Tensor = aten::_convolution(%x, %w, %b, %s, %p, %d, %1, %2, %g, %1, %1, %1)
             return (%4))IR";;
 
     // replace matmul + add pattern to linear