Fix: Remove redundant import os statements causing UnboundLocalError

- Removed redundant 'import os' statements inside functions (lines 4522, 4926, 5005)
- Added back missing 'from llama_cpp import Llama' that was accidentally deleted
- Global 'import os' at line 12 is now the only one

This fixes the UnboundLocalError when running --list-cached-models or other CLI options.

coderai

@@ -4520,7 +4520,6 @@ def main():
                 print(f"Detected GGUF image model, loading with llama.cpp...")
                 try:
                     from llama_cpp import Llama
-                    import os
                     
                     model_key = f"image:{original_model_name}"
                     
@@ -4924,7 +4923,7 @@ def main():
                 print(f"Detected GGUF image model, loading with llama.cpp...")
                 try:
                     from llama_cpp import Llama
-                    import os
+                    from llama_cpp import Llama
                     
                     model_key = f"image:{original_model_name}"
                     
@@ -5003,7 +5002,7 @@ def main():
                     print(f"Warning: Failed to pre-load GGUF image model: {e}")
                 try:
                     from llama_cpp import Llama
-                    import os
+                    from llama_cpp import Llama
                     
                     model_key = f"image:{model_name}"