[DML EP] Add QDQ+MatMul fusion into MatMulNBits (microsoft#22114)

### Description
<!-- Describe your changes. -->



### Motivation and Context
<!-- - Why is this change required? What problem does it solve?
- If it fixes an open issue, please link to the issue here. -->

onnxruntime/core/optimizer/qdq_transformer/selectors_actions/qdq_selector_action_transformer.cc

@@ -290,7 +290,7 @@ void DQMatMulToMatMulNBitsRules(SelectorActionRegistry& qdq_selector_action_regi
                                                          p_buffered_tensors);
 
 #if !defined(ORT_MINIMAL_BUILD)
-  std::vector<const char*> providers = {kCpuExecutionProvider, kCudaExecutionProvider};
+  std::vector<const char*> providers = {kCpuExecutionProvider, kCudaExecutionProvider, kDmlExecutionProvider};
   std::unique_ptr<NodeSelector> selector = std::make_unique<QDQ::DQMatMulToMatMulNBitsSelector>(providers);
   qdq_selector_action_registry.RegisterSelectorAndAction(action_name,
                                                          {{"MatMul", {}}},