Improve AI filtering error handling and visibility

- Add listAvailableModels() and validateModel() to openai.ts
- Improve testOpenAIConnection() to test configured model
- Add checkAIStatus endpoint to filtering router
- Add pre-execution AI config check in executeRules
- Improve error messages in AI filtering service (rate limit, quota, etc.)
- Add AI status warning banner on round detail page for filtering rounds

Now admins get clear errors when AI is misconfigured instead of silent flags.

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

src/lib/openai.ts

@@ -66,12 +66,12 @@ export async function isOpenAIConfigured(): Promise<boolean> {
 }
 
 /**
- * Test OpenAI connection
+ * List available models from OpenAI API
  */
-export async function testOpenAIConnection(): Promise<{
+export async function listAvailableModels(): Promise<{
   success: boolean
+  models?: string[]
   error?: string
-  model?: string
 }> {
   try {
     const client = await getOpenAI()
@@ -83,9 +83,90 @@ export async function testOpenAIConnection(): Promise<{
       }
     }
 
-    // Simple test request
+    const response = await client.models.list()
+    const chatModels = response.data
+      .filter((m) => m.id.includes('gpt') || m.id.includes('o1') || m.id.includes('o3') || m.id.includes('o4'))
+      .map((m) => m.id)
+      .sort()
+
+    return {
+      success: true,
+      models: chatModels,
+    }
+  } catch (error) {
+    return {
+      success: false,
+      error: error instanceof Error ? error.message : 'Unknown error',
+    }
+  }
+}
+
+/**
+ * Validate that a specific model is available
+ */
+export async function validateModel(modelId: string): Promise<{
+  valid: boolean
+  error?: string
+}> {
+  try {
+    const client = await getOpenAI()
+
+    if (!client) {
+      return {
+        valid: false,
+        error: 'OpenAI API key not configured',
+      }
+    }
+
+    // Try a minimal completion with the model
+    await client.chat.completions.create({
+      model: modelId,
+      messages: [{ role: 'user', content: 'test' }],
+      max_tokens: 1,
+    })
+
+    return { valid: true }
+  } catch (error) {
+    const message = error instanceof Error ? error.message : 'Unknown error'
+    // Check for specific model errors
+    if (message.includes('does not exist') || message.includes('model_not_found')) {
+      return {
+        valid: false,
+        error: `Model "${modelId}" is not available with your API key`,
+      }
+    }
+    return {
+      valid: false,
+      error: message,
+    }
+  }
+}
+
+/**
+ * Test OpenAI connection with the configured model
+ */
+export async function testOpenAIConnection(): Promise<{
+  success: boolean
+  error?: string
+  model?: string
+  modelTested?: string
+}> {
+  try {
+    const client = await getOpenAI()
+
+    if (!client) {
+      return {
+        success: false,
+        error: 'OpenAI API key not configured',
+      }
+    }
+
+    // Get the configured model
+    const configuredModel = await getConfiguredModel()
+
+    // Test with the configured model
     const response = await client.chat.completions.create({
-      model: 'gpt-4o-mini',
+      model: configuredModel,
       messages: [{ role: 'user', content: 'Hello' }],
       max_tokens: 5,
     })
@@ -93,11 +174,25 @@ export async function testOpenAIConnection(): Promise<{
     return {
       success: true,
       model: response.model,
+      modelTested: configuredModel,
     }
   } catch (error) {
+    const message = error instanceof Error ? error.message : 'Unknown error'
+    const configuredModel = await getConfiguredModel()
+
+    // Check for model-specific errors
+    if (message.includes('does not exist') || message.includes('model_not_found')) {
+      return {
+        success: false,
+        error: `Model "${configuredModel}" is not available. Check Settings → AI to select a valid model.`,
+        modelTested: configuredModel,
+      }
+    }
+
     return {
       success: false,
-      error: error instanceof Error ? error.message : 'Unknown error',
+      error: message,
+      modelTested: configuredModel,
     }
   }
 }