fix: migrate glm-4-flash to glm-4-flash-250414 (model deprecated by Zhipu)

Zhipu AI has deprecated glm-4-flash, causing 404 errors on all chat requests.
Updated all references:
- config: glm-4-flash → glm-4-flash-250414, added glm-z1-flash
- frontend: defaultModel, conversationStore, ChatArea fallback, ModelsAPI

config/chinese-providers.toml

@@ -25,12 +25,19 @@ max_output_tokens = 4096
 supports_streaming = true
 
 [[llm.providers.models]]
-id = "glm-4-flash"
+id = "glm-4-flash-250414"
-alias = "GLM-4-Flash"
+alias = "GLM-4-Flash (免费)"
 context_window = 128000
 max_output_tokens = 4096
 supports_streaming = true
 
+[[llm.providers.models]]
+id = "glm-z1-flash"
+alias = "GLM-Z1-Flash (免费推理)"
+context_window = 128000
+max_output_tokens = 16384
+supports_streaming = true
+
 [[llm.providers.models]]
 id = "glm-4v-plus"
 alias = "GLM-4V-Plus (视觉)"

config/config.toml

@@ -129,7 +129,7 @@ retry_delay = "1s"
 
 [llm.aliases]
 # 智谱 GLM 模型 (使用正确的 API 模型 ID)
-"glm-4-flash" = "zhipu/glm-4-flash"
+"glm-4-flash" = "zhipu/glm-4-flash-250414"
 "glm-4-plus" = "zhipu/glm-4-plus"
 "glm-4.5" = "zhipu/glm-4.5"
 # 其他模型

desktop/src/components/ChatArea.tsx

@@ -85,7 +85,7 @@ export function ChatArea({ compact, onOpenDetail }: { compact?: boolean; onOpenD
     }
     // Fallback: provide common models when no backend is connected
     return [
-      { id: 'glm-4-flash', name: 'GLM-4 Flash', provider: 'zhipu' },
+      { id: 'glm-4-flash-250414', name: 'GLM-4 Flash (免费)', provider: 'zhipu' },
       { id: 'gpt-4o', name: 'GPT-4o', provider: 'openai' },
       { id: 'gpt-4o-mini', name: 'GPT-4o Mini', provider: 'openai' },
       { id: 'deepseek-chat', name: 'DeepSeek V3', provider: 'deepseek' },
@@ -575,7 +575,7 @@ export function ChatArea({ compact, onOpenDetail }: { compact?: boolean; onOpenD
   );
 }
 
-function MessageBubble({ message, setInput }: { message: Message; setInput: (text: string) => void }) {
+function MessageBubble({ message, setInput, onRetry }: { message: Message; setInput: (text: string) => void; onRetry?: () => void }) {
   if (message.role === 'tool') {
     return null;
   }

desktop/src/components/Settings/ModelsAPI.tsx

@@ -188,7 +188,7 @@ export function ModelsAPI() {
   // 表单状态
   const [formData, setFormData] = useState({
     provider: 'zhipu',
-    modelId: 'glm-4-flash',
+    modelId: 'glm-4-flash-250414',
     displayName: '',
     apiKey: '',
     apiProtocol: 'openai' as 'openai' | 'anthropic' | 'custom',
@@ -678,11 +678,11 @@ export function ModelsAPI() {
                   type="text"
                   value={formData.modelId}
                   onChange={(e) => setFormData({ ...formData, modelId: e.target.value })}
-                  placeholder="如：glm-4-flash, glm-4-plus, glm-4.5"
+                  placeholder="如：glm-4-flash-250414, glm-4-plus, glm-4.7"
                   className="w-full px-3 py-2 border border-gray-200 dark:border-gray-600 rounded-lg text-sm bg-white dark:bg-gray-700 text-gray-900 dark:text-white placeholder-gray-400 focus:outline-none focus:ring-2 focus:ring-orange-500"
                 />
                 <p className="text-xs text-gray-400 mt-1">
-                  智谱: glm-4-flash(免费), glm-4-plus, glm-4.5, glm-4.6
+                  智谱: glm-4-flash-250414(免费), glm-4-plus, glm-4.7, glm-z1-flash(免费推理)
                 </p>
               </div>
 

desktop/src/constants/models.ts

@@ -51,11 +51,11 @@ export const PROVIDER_DEFAULTS = {
   },
   zhipu: {
     baseUrl: 'https://open.bigmodel.cn/api/paas/v4',
-    defaultModel: 'glm-4-flash',
+    defaultModel: 'glm-4-flash-250414',
   },
   zhipu_coding: {
     baseUrl: 'https://open.bigmodel.cn/api/coding/paas/v4',
-    defaultModel: 'glm-4-flash',
+    defaultModel: 'glm-4-flash-250414',
   },
   kimi: {
     baseUrl: 'https://api.moonshot.cn/v1',

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',
+  currentModel: 'glm-4-flash-250414',
 
   newConversation: (currentMessages: ChatMessage[]) => {
     const state = get();