Improve streaming response and add system prompt support

- Add configurable initial streaming message
- Support system prompt in API requests
- Fix config key typo (open-webui -> open_webui)
- Add validation for required config values
- Improve error handling for network and API errors
- Set proper timeout for API requests (900s)
- Better logging for rate limit errors

discord_connector/example.config.yml

@@ -9,7 +9,7 @@ allow_dms:        false   # set to true if you want the bot to answer private DM
 
 # ─────────── Open‑WebUI Settings ───────────
 open_webui_url:    "http://your_open-webui_ip_or_domain:port"
-open-webui_api_key: "user_api_key_from_open_webui"
+open_webui_api_key: "user_api_key_from_open_webui"
 model_name:        "model_id_from_open-webui"
 knowledge_base:    "knowledge_base_id_from_open-webui"
 
@@ -20,5 +20,7 @@ tools:
 
 use_streaming: true # Allows to stream the answer to feel more interactive.
 
+streaming_initial_message: "Bitte warte kurz, die Informationen werden gesammelt..."
+
 # optional system prompt (you can leave it empty to use the default one or the systemprompt given in open-webui for the specific model)
 system_prompt:     ""