Refactor logic to skip unreliable tests (#5919)

e2e/playwright/snapshot-tests.spec.ts

@@ -1,6 +1,12 @@
 import { test, expect } from './zoo-test'
 import { secrets } from './secrets'
-import { Paths, doExport, getUtils, settingsToToml } from './test-utils'
+import {
+  Paths,
+  doExport,
+  getUtils,
+  settingsToToml,
+  orRunWhenFullSuiteEnabled,
+} from './test-utils'
 import { Models } from '@kittycad/lib'
 import fsp from 'fs/promises'
 import { spawn } from 'child_process'
@@ -40,7 +46,7 @@ test(
   'exports of each format should work',
   { tag: ['@snapshot', '@skipWin', '@skipMacos'] },
   async ({ page, context, scene, cmdBar, tronApp }) => {
-    test.fixme(process.env.GITHUB_HEAD_REF !== 'all-e2e')
+    test.fixme(orRunWhenFullSuiteEnabled())
     if (!tronApp) {
       fail()
     }
@@ -1093,7 +1099,7 @@ test.describe('Grid visibility', { tag: '@snapshot' }, () => {
 })
 
 test('theme persists', async ({ page, context }) => {
-  test.fixme(process.env.GITHUB_HEAD_REF !== 'all-e2e')
+  test.fixme(orRunWhenFullSuiteEnabled())
   const u = await getUtils(page)
   await context.addInitScript(async () => {
     localStorage.setItem(