fix(copilot): don't inflate completion tokens to 1 for tool-only responses

When streaming providers don't report usage, the fallback estimator
used max(..., 1) for completion tokens. For tool-only responses with
no text output, this artificially charged 1 phantom token.

autogpt_platform/backend/backend/copilot/baseline/service.py

@@ -454,8 +454,8 @@ async def stream_chat_completion_baseline(
             turn_prompt_tokens = max(
                 estimate_token_count(openai_messages, model=config.model), 1
             )
-            turn_completion_tokens = max(
-                estimate_token_count_str(assistant_text, model=config.model), 1
+            turn_completion_tokens = estimate_token_count_str(
+                assistant_text, model=config.model
             )
             logger.info(
                 "[Baseline] No streaming usage reported; estimated tokens: "