Exception "Dangling name: extfq0cslxr_adminschemakey" when updating schema extension values

Describe the bug

I referred to the Microsoft documentation(https://learn.microsoft.com/en-us/graph/extensibility-overview?tabs=java) on schema extensions and tried using the provided code to update schema extension values directly. However, I encountered an exception with the message "Dangling name: extfq0cslxr_adminschemakey".

I am seeking a way to improve performance and avoid the exception by using direct filters to retrieve and update schema extension values.

Expected behavior

The schema extension values should be updated without any exceptions.

How to reproduce

Steps to Reproduce:

• Use the Microsoft Graph dependency with version 6.x.
• Refer to the following document to update the schema extension value on a particular group.
  -Use the provided Java code:

GraphServiceClient graphClient = new GraphServiceClient(requestAdapter);

User user = new User();
HashMap<String, Object> additionalData = new HashMap<String, Object>();
extkmpdyld2GraphLearnCourses = new ();
extkmpdyld2GraphLearnCourses.setCourseType("Instructor-led");
extkmpdyld2GraphLearnCourses.setCourseId(null);
additionalData.put("extkmpdyld2_graphLearnCourses", extkmpdyld2GraphLearnCourses);
user.setAdditionalData(additionalData);
User result = graphClient.users().byUserId("{user-id}").patch(user);

Encounter the following exception message while updating the value: "Dangling name: extfq0cqexr_adminschema".

SDK Version

6.10.0

Latest version known to work for scenario above?

5.x

Known Workarounds

Below Code works :
GraphServiceClient graphClient = GraphClientBuilder.getGraphClient();
       Map<String, UntypedNode> map = new HashMap<>();
       Map<String, Object> additionalData = new HashMap<>();
       UntypedString productId = new UntypedString(group.getProductId());
       UntypedString productAdminGroupId = new UntypedString("xxx-xx");
       UntypedString linkedOrgGroupId = new UntypedString("xxx-xx-xxx);
       UntypedBoolean adminGroup = new UntypedBoolean(true);
       map.put("productId", productId);
       map.put("productAdminGroupId", productAdminGroupId);
       map.put("linkedOrgGroupId", linkedOrgGroupId);
       map.put("adminGroup", adminGroup);
       UntypedObject untypedObject = new UntypedObject(map);
       additionalData.put(DIConstants.GROUP_ADDITIONAL_ATTRIBUTES_SCHEMA_KEY, untypedObject);
       updateGroup.setAdditionalData(additionalData);
       graphClient.groups().byGroupId(groupId).patch(updateGroup);

Debug output

</details>


### Configuration

_No response_

### Other information

I'm currently using the workaround code,but the following code is impacting the performance of APIs.So i'm looking for the direct filters from graph SDK

首先，使用 Microsoft Graph SDK 6.10.0 重現 Microsoft 文件中的結構描述擴充更新，並將其與回報中可正常運作的 5.x 行為進行比較。追蹤直接更新時 additionalData 的序列化方式，以及「Dangling name」例外在哪裡引發，然後確認文件所述的更新不使用 UntypedNode workaround 也能成功。