Fix: Improve LLM connectivity, add logging, increase timeout, update docs

.env.example

@@ -12,6 +12,9 @@ OPENAPI_URL=http://localhost:8080/v1
 OPENAPI_API_KEY=
 MODEL_NAME=gpt-4o
 
+# LLM Call Timeout in seconds (increase for large documents)
+LLM_TIMEOUT=120
+
 # Summarization Configuration
 # Characters per chunk when splitting long text
 CHUNK_SIZE=4000