fix(ci): parallelize nightly E2E across browsers to prevent timeout

[luans-qa]

Problem

The nightly E2E job has been hitting the 60-minute job limit every day and cancelling before any tests complete.

Root cause: 60 tests (20 per browser × 3 browsers) run serially on workers: 1. With retries: 2 and transfer tests that individually time out at 2 minutes, the full suite needs ~3 hours — 3× the job limit.

Timeline from today's cancelled run (26626516698):

• 08:26:19 — "Running 60 tests using 1 worker"
• 08:26:55 — first test fails, triggers retry cycle
• 08:33:51 — solana transfer timeout (2 min) consumes retry budget
• 09:21:32 — job cancelled after exactly 60 min with ~17/60 tests completed

Fix

Split the single job into a matrix of 3 parallel browser jobs (chromium, firefox, webkit). Each job runs only its 20 tests and has a 90-minute limit, which it comfortably fits within.

Before	After
1 job, 60 serial tests, 60 min limit	3 parallel jobs, 20 tests each, 90 min limit
Cancelled every day before finishing	Each browser completes independently
One timed-out Slack notification	One combined Slack message with per-browser breakdown

Other improvements

• Install only the needed browser per job (playwright install --with-deps ${{ matrix.browser }}) instead of all three — saves ~2 min of setup per job
• fail-fast: false so a webkit failure doesn't cancel the chromium and firefox jobs
• Separate notify job aggregates results from all 3 artifacts into a single Slack message

Test plan

• Trigger via workflow_dispatch and confirm 3 parallel jobs appear
• Confirm all 3 jobs complete within their 90-minute limit
• Confirm a single combined Slack notification fires after all 3 finish

[changeset-bot]

⚠️ No Changeset found

Latest commit: 93d4139

Merging this PR will not cause a version bump for any packages. If these changes should not result in a new version, you're good to go. If these changes should result in a version bump, you need to add a changeset.

This PR includes no changesets

When changesets are added to this PR, you'll see the packages that this PR includes changesets for and the associated semver types

Click here to learn what changesets are, and how to add one.

Click here if you're a maintainer who wants to add a changeset to this PR

[greptile-apps]

_{Reviews (1): Last reviewed commit: "fix(ci): parallelize nightly E2E across ..." | Re-trigger Greptile}

[greptile-apps]

Prompt To Fix All With AI

Fix the following 1 code review issue. Work through them one at a time, proposing concise fixes.

---

### Issue 1 of 1
.github/workflows/e2e-regression-tests.yml:212-217
When a browser artifact download fails (silently, due to `continue-on-error: true`) but the matrix job itself succeeded, `result` will be `null` for that browser. The loop only sets `overallFailed = true` when `result && result.failedTests > 0`, so a missing artifact from an otherwise-healthy matrix run would leave `overallFailed` as `false`. The Slack message would then show "✅ Passed" alongside a "No results" line — a misleading green notification.

```suggestion
                  for (const browser of browsers) {
                      const filePath = path.join('test-results', browser, 'playwright-results.json');
                      const result = parseResults(filePath);
                      browserResults[browser] = result;
                      if (!result) overallFailed = true;
                      if (result && result.failedTests > 0) overallFailed = true;
                  }
```

_{Reviews (2): Last reviewed commit: "fix(ci): improve Slack notification form..." | Re-trigger Greptile}

[luans-qa]

Done — extracted to apps/wallets/quickstart-devkit/scripts/notify-slack.js, alongside the existing generate-test-report.js. The workflow step is now a single node apps/wallets/quickstart-devkit/scripts/notify-slack.js call (c62c5d2).

[greptile-apps]

Prompt To Fix All With AI

Fix the following 1 code review issue. Work through them one at a time, proposing concise fixes.

---

### Issue 1 of 1
apps/wallets/quickstart-devkit/scripts/notify-slack.js:78-82
When tests are stored directly on `suite.tests` (rather than in `suite.specs`), the failure record uses `suite: parentPath || ""` — which is the path *before* the current suite, so the current suite's own title is dropped from the failure context shown in Slack. For any test nested directly under a describe block, the suite field will be truncated or empty. Using `currentPath` here aligns with how specs are handled.

```suggestion
                        failures.push({
                            title: test.title || suite.title || "Unknown Test",
                            suite: currentPath,
                            error: result?.error?.message || "No error message",
                        });
```

_{Reviews (3): Last reviewed commit: "refactor(ci): extract Slack notify scrip..." | Re-trigger Greptile}

[github-actions]

🔥 Smoke Test Results

❌ Status: Failed

Statistics

• Total Tests: 5
• Passed: 0 ✅
• Failed: 1 ❌
• Skipped: 4 ⚠️
• Duration: 1.70 min

Test Details

---

This is a non-blocking smoke test. Full regression tests run separately.

[greptile-apps]

_{Reviews (4): Last reviewed commit: "fix(ci): apply Biome formatting to notif..." | Re-trigger Greptile}