fix: clean up deprecated code

Author: ashwinb

src/llama_stack_client/lib/inference/event_logger.py

@@ -5,7 +5,7 @@
 # the root directory of this source tree.
 from typing import Generator
 from termcolor import cprint
-from llama_stack_client.types import ChatCompletionResponseStreamChunk, ChatCompletionChunk
+from llama_stack_client.types import ChatCompletionChunk
 
 
 class InferenceStreamPrintableEvent:
@@ -28,35 +28,11 @@ def __init__(self):
         self.is_thinking = False
 
     def yield_printable_events(
-        self, chunk: ChatCompletionResponseStreamChunk | ChatCompletionChunk
+        self, chunk: ChatCompletionChunk
     ) -> Generator[InferenceStreamPrintableEvent, None, None]:
-        # Check if the chunk has event attribute (ChatCompletionResponseStreamChunk)
-        if hasattr(chunk, "event"):
-            yield from self._handle_inference_stream_chunk(chunk)
-        # Check if the chunk has choices attribute (ChatCompletionChunk)
-        elif hasattr(chunk, "choices") and len(chunk.choices) > 0:
+        if hasattr(chunk, "choices") and len(chunk.choices) > 0:
             yield from self._handle_chat_completion_chunk(chunk)
 
-    def _handle_inference_stream_chunk(
-        self, chunk: ChatCompletionResponseStreamChunk
-    ) -> Generator[InferenceStreamPrintableEvent, None, None]:
-        event = chunk.event
-        if event.event_type == "start":
-            yield InferenceStreamPrintableEvent("Assistant> ", color="cyan", end="")
-        elif event.event_type == "progress":
-            if event.delta.type == "reasoning":
-                if not self.is_thinking:
-                    yield InferenceStreamPrintableEvent("<thinking> ", color="magenta", end="")
-                    self.is_thinking = True
-                yield InferenceStreamPrintableEvent(event.delta.reasoning, color="magenta", end="")
-            else:
-                if self.is_thinking:
-                    yield InferenceStreamPrintableEvent("</thinking>", color="magenta", end="")
-                    self.is_thinking = False
-                yield InferenceStreamPrintableEvent(event.delta.text, color="yellow", end="")
-        elif event.event_type == "complete":
-            yield InferenceStreamPrintableEvent("")
-
     def _handle_chat_completion_chunk(
         self, chunk: ChatCompletionChunk
     ) -> Generator[InferenceStreamPrintableEvent, None, None]: