e2e improvements

Author: gallayl

.cursor/rules/TESTING_GUIDELINES.md

@@ -287,6 +287,95 @@ const errorNoty = page.locator('shade-noty').first()
 await expect(errorNoty).toBeVisible()
 ```
 
+### User Journey Test Pattern
+
+E2E tests should follow user journeys, not component isolation. Each test should simulate a complete user workflow from start to finish.
+
+**Structure:**
+
+- One test per complete user workflow
+- Test from login to logical endpoint
+- Clean up any created/modified data at the end
+- Use helper functions for reusable steps within the test file
+
+**Good Example - User Journey:**
+
+```typescript
+test('Admin can navigate to users, edit roles, and verify persistence', async ({ page }) => {
+  // 1. Login as admin
+  await login(page)
+
+  // 2. Navigate to app settings, verify Users menu
+  await navigateToAppSettings(page)
+  await expect(page.getByText('Users')).toBeVisible()
+
+  // 3. Click Users, verify table structure
+  await page.getByText('Users').click()
+  await verifyUsersTableStructure(page)
+
+  // 4. Open user, record initial state
+  const initialRoleCount = await page.locator('role-tag').count()
+
+  // 5. Make changes (add a role)
+  await addRoleToUser(page)
+  await page.getByRole('button', { name: 'Save' }).click()
+
+  // 6. Reload and verify persistence
+  await page.reload()
+  await expect(page.locator('role-tag')).toHaveCount(initialRoleCount + 1)
+
+  // 7. Cleanup - restore original state
+  await removeRoleFromUser(page)
+  await page.getByRole('button', { name: 'Save' }).click()
+  await expect(page.locator('role-tag')).toHaveCount(initialRoleCount)
+})
+```
+
+**Bad Example - Fragmented Tests:**
+
+```typescript
+// ❌ Don't split into many small tests - causes repeated setup and state issues
+test.describe('User Management', () => {
+  test('should display Users menu', ...)
+  test('should navigate to users list', ...)
+  test('should display table headers', ...)
+  test('should display at least one user', ...)
+  test('should open user details', ...)
+  test('should add a role', ...)
+  test('should save changes', ...)
+})
+```
+
+**Cleanup Pattern:**
+
+```typescript
+test('User can create, customize, and delete a resource', async ({ page }) => {
+  await login(page)
+
+  // Record initial state if needed
+  const initialCount = await page.locator('.resource').count()
+
+  // Create resource (use unique identifier)
+  const resourceName = `test-${Date.now()}`
+  await createResource(page, resourceName)
+
+  // Perform actions and verifications
+  await customizeResource(page, resourceName)
+  await verifyResource(page, resourceName)
+
+  // Cleanup - restore original state
+  await deleteResource(page, resourceName)
+  await expect(page.locator('.resource')).toHaveCount(initialCount)
+})
+```
+
+**Key Principles:**
+
+- Tests are self-contained and don't depend on other tests
+- Tests clean up after themselves to avoid state accumulation
+- Use unique identifiers (timestamps, UUIDs) for test data
+- Helper functions should be local to the test file unless truly reusable across multiple test files
+
 ## Unit Testing Best Practices
 
 ### Component Testing