feat(error-handling): implement LLM error adaptation and classification for chat streaming

- Introduced LLMErrorCategory and adapt_llm_exception to normalize LLM exceptions.
- Updated llm_retryable_message and llm_permanent_message to utilize the new adaptation logic.
- Enhanced classify_stream_exception to classify provider errors and return user-friendly messages.
- Added tests for error classification and adaptation to ensure robustness.
- Updated frontend error handling to display appropriate messages based on new classifications.

surfsense_web/app/dashboard/[search_space_id]/new-chat/[[...chat_id]]/page.tsx

@@ -613,6 +613,18 @@ export default function NewChatPage() {
 				return;
 			}
 
+			if (normalized.channel === "inline") {
+				if (normalized.assistantMessage) {
+					await persistAssistantErrorMessage({
+						threadId,
+						assistantMsgId,
+						text: normalized.assistantMessage,
+					});
+				}
+				toast.error(normalized.userMessage);
+				return;
+			}
+
 			toast.error(normalized.userMessage);
 		},
 		[currentUser?.id, persistAssistantErrorMessage, searchSpaceId, setPremiumAlertForThread]