fix(nlp): remove options n_threads=os.cpu_count() in load_model_nlp

app/routes/nlp.py

@@ -22,7 +22,6 @@ def load_model_nlp():
     llm = Llama.from_pretrained(
         repo_id="Qwen/Qwen2.5-1.5B-Instruct-GGUF",
         filename="qwen2.5-1.5b-instruct-q3_k_m.gguf",
-        n_threads=os.cpu_count()
         use_mmap=True,
         verbose=True,
     )