fix(turns): defer LocalSmartTurnAnalyzerV3 import to avoid loading transformers at module load

Importing pipecat.turns.user_turn_strategies pulled in
LocalSmartTurnAnalyzerV3 → transformers → onnxruntime at module load
time. Since this module is imported by llm_response_universal (and
therefore most LLM services), any LLM service import paid the cost of
loading transformers and triggered its missing-backend warning in
environments without PyTorch/TF/Flax.

Move the LocalSmartTurnAnalyzerV3 import into
default_user_turn_stop_strategies() so it only loads when the default
smart-turn strategy is actually constructed.

Fixes #4392

src/pipecat/turns/user_turn_strategies.py

@@ -8,7 +8,6 @@
 
 from dataclasses import dataclass
 
-from pipecat.audio.turn.smart_turn.local_smart_turn_v3 import LocalSmartTurnAnalyzerV3
 from pipecat.turns.user_start import (
     BaseUserTurnStartStrategy,
     ExternalUserTurnStartStrategy,
@@ -44,6 +43,8 @@ def default_user_turn_stop_strategies() -> list[BaseUserTurnStopStrategy]:
     Returns ``[TurnAnalyzerUserTurnStopStrategy(LocalSmartTurnAnalyzerV3)]``.
     Useful when building a custom strategy list that extends the defaults.
     """
+    from pipecat.audio.turn.smart_turn.local_smart_turn_v3 import LocalSmartTurnAnalyzerV3
+
     return [TurnAnalyzerUserTurnStopStrategy(turn_analyzer=LocalSmartTurnAnalyzerV3())]