fix: add user-visible error handling for spec creation agent start

The onComplete handler in the empty Kanban spec creation flow only
logged errors to console.error, leaving users with no feedback when
the agent failed to start. This wires up the SpecCreationChat
component's built-in error UI (spinner, error banner, retry button)
via initializerStatus, initializerError, and onRetryInitializer props.

Changes:
- Add InitializerStatus type and specInitializerStatus/Error state
- Set status to 'starting' before startAgent call (shows spinner)
- On error, keep spec chat open so the error banner is visible
- On success, close chat and refresh queries (same as before)
- Wire up onRetryInitializer to reset state to idle
- Reset initializer status on cancel/exit for clean re-entry

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

ui/src/App.tsx

@@ -41,6 +41,8 @@ const VIEW_MODE_KEY = 'autocoder-view-mode'
 // Bottom padding for main content when debug panel is collapsed (40px header + 8px margin)
 const COLLAPSED_DEBUG_PANEL_CLEARANCE = 48
 
+type InitializerStatus = 'idle' | 'starting' | 'error'
+
 function App() {
   // Initialize selected project from localStorage
   const [selectedProject, setSelectedProject] = useState<string | null>(() => {
@@ -63,6 +65,8 @@ function App() {
   const [isSpecCreating, setIsSpecCreating] = useState(false)
   const [showResetModal, setShowResetModal] = useState(false)
   const [showSpecChat, setShowSpecChat] = useState(false)  // For "Create Spec" button in empty kanban
+  const [specInitializerStatus, setSpecInitializerStatus] = useState<InitializerStatus>('idle')
+  const [specInitializerError, setSpecInitializerError] = useState<string | null>(null)
   const [viewMode, setViewMode] = useState<ViewMode>(() => {
     try {
       const stored = localStorage.getItem(VIEW_MODE_KEY)
@@ -496,22 +500,30 @@ function App() {
           <SpecCreationChat
             projectName={selectedProject}
             onComplete={async (_specPath, yoloMode) => {
-              // Auto-start the agent after spec creation (same as NewProjectModal)
+              setSpecInitializerStatus('starting')
               try {
                 await startAgent(selectedProject, {
                   yoloMode: yoloMode ?? false,
                   maxConcurrency: 3,
                 })
+                // Success — close chat and refresh
+                setShowSpecChat(false)
+                setSpecInitializerStatus('idle')
+                queryClient.invalidateQueries({ queryKey: ['projects'] })
+                queryClient.invalidateQueries({ queryKey: ['features', selectedProject] })
               } catch (err) {
-                console.error('Failed to start agent:', err)
+                setSpecInitializerStatus('error')
+                setSpecInitializerError(err instanceof Error ? err.message : 'Failed to start agent')
               }
-              setShowSpecChat(false)
-              // Refresh projects to update has_spec
-              queryClient.invalidateQueries({ queryKey: ['projects'] })
-              queryClient.invalidateQueries({ queryKey: ['features', selectedProject] })
             }}
-            onCancel={() => setShowSpecChat(false)}
-            onExitToProject={() => setShowSpecChat(false)}
+            onCancel={() => { setShowSpecChat(false); setSpecInitializerStatus('idle') }}
+            onExitToProject={() => { setShowSpecChat(false); setSpecInitializerStatus('idle') }}
+            initializerStatus={specInitializerStatus}
+            initializerError={specInitializerError}
+            onRetryInitializer={() => {
+              setSpecInitializerError(null)
+              setSpecInitializerStatus('idle')
+            }}
           />
         </div>
       )}