fix(ci): stabilize BrowserStack Test workflow

[crisryantan]

Summary

Makes the BrowserStack Test workflow reliably green on main. This is not caused by a recent product PR: master run 26452312354 (May 26) and PR run 31624825318 fail the same way.

Failure modes in the logs:

1. Edge 15.15063 (EdgeHTML) deterministically failed ~47 of ~1050 tests. Dropped from the matrix (review: Edge 15–18 is ~0.02% share). Chromium Edge betas in the beta workflow are unchanged. No src/ changes — EdgeHTML workarounds were reverted after dropping the browser.
2. Occasional 6-hour hangs until GitHub's default job timeout cancels the run.
3. BrowserStack tunnel flake — concurrent runs hijacking tunnels (transport error, has not captured in 60000 ms, Rate Limit Exceeded).
4. Firefox timing flakes on remaining browsers — Firefox 51 (legacy matrix) and Firefox 153 (beta) still needed a few test waits after Edge was gone.

Changes

• test/cross-browser-testing/ — remove bs_edge_windows_15; share karma settings; pin sessions to BROWSERSTACK_LOCAL_IDENTIFIER; raise capture / no-activity / mocha timeouts.
• .github/workflows/cross-browser-testing.yml — timeout-minutes: 30 and a shared concurrency group for push + pull_request.
• test/src/config/utils.js — raise waitForCondition default timeout for slow BrowserStack VMs.
• test/src/tests-core-sdk.js — wait for in-flight identify before retrying in the failed-config spec (Firefox 51).
• test/src/tests-mparticle-instance-manager.ts — wait for all instance cookies before asserting (Firefox 153 beta).

Coverage tradeoff

Legacy EdgeHTML is no longer in the BrowserStack matrix (~0.02% share). Consent / cookie-sync coverage is not skipped on remaining browsers.

Test plan

• npm run lint clean
• npm run test:jest — 589/589
• BrowserStack on this branch (run 31711843399) — 5344 SUCCESS in ~3.5 min, including Edge 15 (before it was dropped)
• BrowserStack after dropping Edge 15 — remaining matrix green within 30 min
• BrowserStack Beta — green after Firefox 153 cookie wait

[cursor]

PR Summary

Low Risk
Changes are limited to CI workflows, Karma test infrastructure, and test timing/assertions; no production SDK behavior is modified in the provided diff.

Overview
Stabilizes BrowserStack Test by capping jobs at 30 minutes, adding a concurrency group so push and PR on the same branch do not run two matrices at once, and applying the same timeout on the beta workflow.

Karma / BrowserStack: Duplicated config moves into browserstack.karma.shared.js (getSharedKarmaSettings). When BROWSERSTACK_LOCAL_IDENTIFIER is set, Karma uses the CI tunnel (startTunnel: false). Shared settings raise Mocha timeout to 20s; the main matrix also raises capture and browser no-activity timeouts. Edge 15 (EdgeHTML) is removed from the legacy matrix (~0.02% share).

Tests: waitForCondition default timeout increases from 200ms to 3s. Core SDK and multi-instance specs wait for the right async state (in-flight identify, per-instance cookies) instead of assuming fast VMs.

^{Reviewed by Cursor Bugbot for commit 54cef9e. Bugbot is set up for automated code reviews on this repo. Configure here.}

[sonarqubecloud]

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud

[cursor]

Cursor Bugbot has reviewed your changes using default effort and found 1 potential issue.

^{❌ Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, have a team admin enable autofix in the Cursor dashboard.}

^{Reviewed by Cursor Bugbot for commit baf751e. Configure here.}

[rmi22186]

Edge 15–18 marketshare is only .02%. let's just drop edge from our browserstack

[crisryantan]

Dropped bs_edge_windows_15 from the BrowserStack matrix in 2e5f396 (Edge 15–18 ~0.02% share). Chromium Edge in the beta workflow is unchanged. The performCookieSync try/catch stays as production hardening for invalid pixel URLs.

[sonarqubecloud]

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud