feat(llm): add NVIDIA NIM, Kimi, Groq, and Mistral providers

Each is a one-row entry in the OpenAI-compatible provider registry (base_url,
key env, CLI option); the model is user-specified since they serve many models.

.env.example

@@ -11,6 +11,15 @@ ZHIPU_CN_API_KEY=
 MINIMAX_API_KEY=
 MINIMAX_CN_API_KEY=
 OPENROUTER_API_KEY=
+MISTRAL_API_KEY=
+MOONSHOT_API_KEY=
+GROQ_API_KEY=
+NVIDIA_API_KEY=
+
+# Optional: a custom OpenAI-compatible endpoint (vLLM, LM Studio, llama.cpp,
+# relay). Select provider "openai_compatible" and set the base URL; the key is
+# optional (local servers need none).
+#OPENAI_COMPATIBLE_API_KEY=
 
 # Optional: point at a remote Ollama server. When unset, defaults to
 # the local instance at http://localhost:11434/v1. Convention follows