Python: pseudo-stream copilot invoke_stream (#12548)

### Motivation and Context

The current copilot studio client doesn't support streaming invocation.
The underlying orchestration patterns rely on agents to have an
invoke_stream. Since it doesn't exist in the copilot agent, that agent
type cannot be used. This PR implements a pseudo-invoke_stream so that
the agent can be used with patterns - although it doesn't stream the
response back in chunks, it unblocks the use if one desired to use our
new orchestration patterns.

<!-- Thank you for your contribution to the semantic-kernel repo!
Please help reviewers and future users, providing the following
information:
  1. Why is this change required?
  2. What problem does it solve?
  3. What scenario does it contribute to?
  4. If it fixes an open issue, please link to the issue here.
-->

### Description

- Closes #12449

<!-- Describe your changes, the overall approach, the underlying design.
These notes will help understanding how your code works. Thanks! -->

### Contribution Checklist

<!-- Before submitting this PR, please make sure: -->

- [X] The code builds clean without any errors or warnings
- [X] The PR follows the [SK Contribution
Guidelines](https://github.com/microsoft/semantic-kernel/blob/main/CONTRIBUTING.md)
and the [pre-submission formatting
script](https://github.com/microsoft/semantic-kernel/blob/main/CONTRIBUTING.md#development-scripts)
raises no violations
- [X] All unit tests pass, and I have added new tests where possible
- [X] I didn't break anyone 😄

python/semantic_kernel/agents/copilot_studio/copilot_studio_agent.py

@@ -32,6 +32,7 @@ from semantic_kernel.agents.copilot_studio.copilot_studio_agent_settings import
     CopilotStudioAgentSettings,
 )
 from semantic_kernel.contents.chat_message_content import ChatMessageContent
+from semantic_kernel.contents.streaming_chat_message_content import StreamingChatMessageContent
 from semantic_kernel.contents.utils.author_role import AuthorRole
 from semantic_kernel.exceptions.agent_exceptions import (
     AgentInitializationException,
@@ -54,7 +55,6 @@ else:  # pragma: no cover
     from typing_extensions import override
 
 if TYPE_CHECKING:  # pragma: no cover
-    from semantic_kernel.contents.streaming_chat_message_content import StreamingChatMessageContent
     from semantic_kernel.kernel import Kernel
 
 logger: logging.Logger = logging.getLogger(__name__)
@@ -510,7 +510,7 @@ class CopilotStudioAgent(Agent):
             yield AgentResponseItem(message=response, thread=thread)
 
     @override
-    def invoke_stream(
+    async def invoke_stream(
         self,
         messages: str | ChatMessageContent | list[str | ChatMessageContent] | None = None,
         *,
@@ -519,8 +519,53 @@ class CopilotStudioAgent(Agent):
         arguments: KernelArguments | None = None,
         kernel: "Kernel | None" = None,
         **kwargs: Any,
-    ) -> AsyncIterable[AgentResponseItem["StreamingChatMessageContent"]]:
-        raise NotImplementedError("Streaming is not supported for Copilot Studio agents.")
+    ) -> AsyncIterable[AgentResponseItem[StreamingChatMessageContent]]:
+        """Invoke the agent and stream the response.
+
+        Note: this is a “pseudo-streaming” implementation.
+
+        We're internally delegating to the real async generator `_inner_invoke`.
+        Each complete ChatMessageContent is wrapped in exactly one
+        StreamingChatMessageContent chunk, so downstream consumers can iterate
+        without change.  The stream yields at least once; callers still receive
+        on_intermediate callbacks in real time.
+
+        Args:
+            messages: The messages to send to the agent.
+            thread: The thread to use for the agent.
+            on_intermediate_message: A callback function to call with each intermediate message.
+            arguments: The arguments to pass to the agent.
+            kernel: The kernel to use for the agent.
+            **kwargs: Additional keyword arguments.
+
+        Yields:
+            A chat message content and thread with the response.
+        """
+        thread = await self._ensure_thread_exists_with_messages(
+            messages=messages,
+            thread=thread,
+            construct_thread=lambda: CopilotStudioAgentThread(self.client),
+            expected_type=CopilotStudioAgentThread,
+        )
+        if not isinstance(thread, CopilotStudioAgentThread):
+            raise AgentThreadOperationException("The thread is not a Copilot Studio Agent thread.")
+
+        normalized_messages = self._normalize_messages(messages)
+
+        responses: list[ChatMessageContent] = []
+        async for resp in self._inner_invoke(
+            thread=thread,
+            messages=normalized_messages,
+            on_intermediate_message=on_intermediate_message,
+            arguments=arguments,
+            kernel=kernel,
+            **kwargs,
+        ):
+            responses.append(resp)
+
+        for i, resp in enumerate(responses):
+            stream_msg = self._to_streaming(resp, index=i)
+            yield AgentResponseItem(message=stream_msg, thread=thread)
 
     # endregion
 
@@ -598,6 +643,21 @@ class CopilotStudioAgent(Agent):
             normalized.append(m.content if isinstance(m, ChatMessageContent) else str(m))
         return normalized
 
+    @staticmethod
+    def _to_streaming(
+        msg: ChatMessageContent,
+        *,
+        index: int,
+    ) -> StreamingChatMessageContent:
+        """Wrap a complete ChatMessageContent in a StreamingChatMessageContent."""
+        return StreamingChatMessageContent(
+            role=msg.role,
+            name=msg.name,
+            content=msg.content,
+            choice_index=index,
+            metadata=msg.metadata,
+        )
+
     @override
     async def _notify_thread_of_new_message(self, thread, new_message):
         """Copilot Studio Agent doesn't need to notify the thread of new messages.