Refactor configuration and assessment modules; add S3 integration and update LLM client usage

example.run_config.yaml

@@ -0,0 +1,39 @@
+# Helia Run Configuration
+# This file defines the "providers" (LLM connections) and the "runs" (experiments).
+# Environment variables like ${OPENAI_API_KEY} are expanded at runtime.
+
+providers:
+  openai:
+    api_key: "${OPENAI_API_KEY}"
+    api_base: "https://api.openai.com/v1"
+    api_format: "openai"
+
+  anthropic:
+    api_key: "${ANTHROPIC_API_KEY}"
+    api_base: "https://api.anthropic.com/v1"
+    api_format: "anthropic"
+
+  openrouter:
+    api_key: "${OPENROUTER_API_KEY}"
+    api_base: "https://openrouter.ai/api/v1"
+    api_format: "openai"
+
+  local_ollama:
+    api_key: "none"
+    api_base: "http://localhost:11434/v1"
+    api_format: "ollama"
+
+runs:
+  - run_name: "baseline_gpt4"
+    model:
+      provider: openai
+      model_name: "gpt-4o"
+      temperature: 0.0
+    prompt_id: "default"
+
+  - run_name: "test_llama3"
+    model:
+      provider: local_ollama
+      model_name: "llama3"
+      temperature: 0.7
+    prompt_id: "default"