Fold BaseTask.handoff_to into activate_task(deactivate_self=...)

BaseTask.handoff_to was just deactivate_self + activate_task. Remove
it and add a deactivate_self flag on activate_task instead, so there's
one entry point for activating another task.

LLMTask now overrides activate_task (mirroring its end() override) to
keep the messages / result_callback hooks that finish an in-progress
tool call before the target is activated. All multi-task examples and
unit tests switch to the new call.

examples/multi-task/local-handoff/local-handoff-two-agents-tts.py

@@ -106,7 +106,7 @@ class AcmeTTSTask(LLMTask):
             reason (str): Why the user is being transferred.
         """
         logger.info(f"Task '{self.name}': transferring to '{agent}' ({reason})")
-        await self.handoff_to(
+        await self.activate_task(
             agent,
             messages=[
                 {
@@ -114,9 +114,10 @@ class AcmeTTSTask(LLMTask):
                     "content": f"Tell the user about the transfer ({reason}).",
                 }
             ],
-            activation_args=LLMTaskActivationArgs(
+            args=LLMTaskActivationArgs(
                 messages=[{"role": "developer", "content": reason}],
             ),
+            deactivate_self=True,
             result_callback=params.result_callback,
         )
 

examples/multi-task/local-handoff/local-handoff-two-agents.py

@@ -85,11 +85,12 @@ class AcmeLLMTask(LLMTask):
             reason (str): Why the user is being transferred.
         """
         logger.info(f"Task '{self.name}': transferring to '{agent}' ({reason})")
-        await self.handoff_to(
+        await self.activate_task(
             agent,
-            activation_args=LLMTaskActivationArgs(
+            args=LLMTaskActivationArgs(
                 messages=[{"role": "developer", "content": reason}],
             ),
+            deactivate_self=True,
             result_callback=params.result_callback,
         )