Fix critical model name issue and validator warning

Critical fix for guarded_ai.py:
- Add MODEL_NAME_{n} environment variable support
- Fixes hard-coded "model" string that breaks Azure OpenAI and other endpoints
- Falls back to "model" if MODEL_NAME_{n} not specified
- Some endpoints require actual deployment name in model parameter

Validator improvement:
- Allow feedback_prompts as alternative to feedback_tokens_for_ai
- Prevents false warning when using metadata_feedback_filter with new prompt system

Documentation:
- Added MODEL_NAME_{n} examples to CLAUDE.md
- Documented that Azure and similar endpoints need this variable

All 8 activities validated: 0 errors, 0 warnings

research/guarded_ai.py

@@ -48,14 +48,15 @@ def get_openai_client_and_model(model_name=None):
             model_num = model_name.split("_")[1]
             endpoint_key = f"MODEL_ENDPOINT_{model_num}"
             api_key_key = f"MODEL_API_KEY_{model_num}"
+            model_name_key = f"MODEL_NAME_{model_num}"
 
             endpoint = os.getenv(endpoint_key)
             api_key = os.getenv(api_key_key)
 
             if endpoint and api_key:
                 client = get_client_for_endpoint(endpoint, api_key)
-                # Use a simple default model name for the endpoint
-                actual_model = "model"  # Most endpoints use "model" or ignore this
+                # Get the actual model name from environment, or use sensible default
+                actual_model = os.getenv(model_name_key) or "model"
                 return client, actual_model
         except Exception as e:
             print(f"Warning: Failed to load {model_name}: {e}, falling back to default")