[VitisAI] change all support tensor type from ir 9 to ir 10 (#23204)

### Description
<!-- Describe your changes. -->
Changed all support tensor  type from ir 9 to ir 10.


### 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. -->
- See issue #23205

Co-authored-by: Yueqing Zhang <[email protected]>

include/onnxruntime/core/framework/data_types.h

@@ -208,6 +208,7 @@ class DataTypeImpl {
   static const std::vector<MLDataType>& AllTensorTypes();  // up to IR4 (no float 8), deprecated
   static const std::vector<MLDataType>& AllTensorTypesIRv4();
   static const std::vector<MLDataType>& AllTensorTypesIRv9();
+  static const std::vector<MLDataType>& AllTensorTypesIRv10();
 
   static const std::vector<MLDataType>& AllFixedSizeTensorTypes();  // up to IR4 (no float 8), deprecated
   static const std::vector<MLDataType>& AllFixedSizeTensorTypesIRv4();

onnxruntime/core/framework/data_types.cc

@@ -1301,6 +1301,12 @@ const std::vector<MLDataType>& DataTypeImpl::AllTensorTypesIRv9() {
   return all_tensor_types;
 }
 
+const std::vector<MLDataType>& DataTypeImpl::AllTensorTypesIRv10() {
+  static std::vector<MLDataType> all_tensor_types =
+      GetTensorTypesFromTypeList<element_type_lists::AllIRv10>();
+  return all_tensor_types;
+}
+
 const std::vector<MLDataType>& DataTypeImpl::AllFixedSizeSequenceTensorTypes() {
   return AllFixedSizeSequenceTensorTypesIRv4();
 }

onnxruntime/core/session/custom_ops.cc

@@ -49,7 +49,7 @@ static constexpr uint32_t min_ort_version_with_shape_inference = 17;
 #endif
 
 #if !defined(DISABLE_FLOAT8_TYPES)
-#define SUPPORTED_TENSOR_TYPES DataTypeImpl::AllTensorTypesIRv9()
+#define SUPPORTED_TENSOR_TYPES DataTypeImpl::AllTensorTypesIRv10()
 #else
 #define SUPPORTED_TENSOR_TYPES DataTypeImpl::AllTensorTypesIRv4()
 #endif