Update _llm_response_generation.py

Update llamacpp.py

Fix llama-server output piping (handle invalid utf8 chars gracefully)

Update model_metadata.py

fix llama-server cli args

add missing model metadata for llama.cpp models, add phi-4-mini-instruct

update gguf models

fix LlamaCppHandler

preserve model_name further

Add some model mappings to LlamaCppHandler

[BFCL] Add New Model DeepSeek-R1 (ShishirPatil#901)