add qwen3-vl, fix librechat taking over settings and clean up llama config

apps/librechat/release.yaml

@@ -57,7 +57,9 @@ spec:
                   "Qwen3-4B-Thinking-2507-long-ctx",
                   "Qwen2.5-VL-7B-Instruct-GGUF",
                   "Qwen2.5-VL-32B-Instruct-GGUF-IQ1_S",
-                  "Qwen2.5-VL-32B-Instruct-GGUF-Q2_K_L"
+                  "Qwen2.5-VL-32B-Instruct-GGUF-Q2_K_L",
+                  "Qwen3-VL-4B-Instruct-GGUF",
+                  "Qwen3-VL-4B-Instruct-GGUF-unslothish"
                 ]
               titleConvo: true
               titleModel: "gemma3-4b-novision"
@@ -65,6 +67,16 @@ spec:
               summaryModel: "gemma3-4b-novision"
               forcePrompt: false
               modelDisplayLabel: "Llama.cpp"
+
+              # ✨ IMPORTANT: let llama-swap/llama-server own all these
+              dropParams:
+                - "temperature"
+                - "top_p"
+                - "top_k"
+                - "presence_penalty"
+                - "frequency_penalty"
+                - "stop"
+                - "max_tokens"
       imageVolume:
         enabled: true
         size: 10G