Fix output final text for HuggingFaceTextGenInference when streaming (#6211)

The LLM integration
[HuggingFaceTextGenInference](https://github.com/hwchase17/langchain/blob/master/langchain/llms/huggingface_text_gen_inference.py)
already has streaming support.

However, when streaming is enabled, it always returns an empty string as
the final output text when the LLM is finished. This is because `text`
is instantiated with an empty string and never updated.

This PR fixes the collection of the final output text by concatenating
new tokens.

langchain/llms/huggingface_text_gen_inference.py

@@ -169,4 +169,5 @@ class HuggingFaceTextGenInference(LLM):
                 if not token.special:
                     if text_callback:
                         text_callback(token.text)
+                    text += token.text
         return text