fix: sync currentModel from SaaS available models on login

Root cause: conversationStore hardcoded 'glm-4-flash' as default model,
which may not exist in SaaS admin config, causing 404 on all chat requests.

- conversationStore: default currentModel to empty string (runtime-resolved)
- saasStore: after fetching available models, auto-switch currentModel
  to first available if the stored model is not in the list
- SaaS relay getModel() already had fallback to first available model

desktop/src/store/chat/conversationStore.ts

@@ -192,7 +192,7 @@ export const useConversationStore = create<ConversationState>()(
   agents: [DEFAULT_AGENT],
   currentAgent: DEFAULT_AGENT,
   sessionKey: null,
-  currentModel: 'glm-4-flash-250414',
+  currentModel: '',  // Set dynamically: SaaS models or config.toml default
 
   newConversation: (currentMessages: ChatMessage[]) => {
     const state = get();

desktop/src/store/saasStore.ts

@@ -467,6 +467,24 @@ export const useSaaSStore = create<SaaSStore>((set, get) => {
         saasClient.setBaseUrl(saasUrl);
         const models = await saasClient.listModels();
         set({ availableModels: models });
+
+        // Sync currentModel: if the stored model is not in the available list,
+        // switch to the first available model to prevent 404 errors
+        if (models.length > 0) {
+          try {
+            const { useConversationStore } = require('./chat/conversationStore');
+            const current = useConversationStore.getState().currentModel;
+            const modelIds = models.map(m => m.alias || m.id);
+            if (current && !modelIds.includes(current)) {
+              const firstModel = models[0];
+              const fallbackId = firstModel.alias || firstModel.id;
+              useConversationStore.getState().setCurrentModel(fallbackId);
+              log.info(`Synced currentModel: ${current} not available, switched to ${fallbackId}`);
+            }
+          } catch (syncErr) {
+            log.warn('Failed to sync currentModel after fetching models:', syncErr);
+          }
+        }
       } catch (err: unknown) {
         log.warn('Failed to fetch available models:', err);
         // Do not set error state - model fetch failure is non-critical