fix(function): getTargetId

[Kikobeats]

Note

Medium Risk
Changes how page-based functions pick the Puppeteer page by adding CDP targetId plumbing, which can affect runtime behavior under load and across different Puppeteer/Chrome versions. Includes a new concurrency test that should catch regressions but still warrants review in real deployments.

Overview
Fixes a concurrency bug where page-based functions could run against the wrong Puppeteer page when multiple pages exist.

The wrapper now captures the current page’s CDP targetId and passes it into the isolated function; the generated template uses that targetId to find the matching page after puppeteer.connect, falling back to the last page if unavailable. Adds a test asserting concurrent requests each receive their own page, and tweaks nano-staged to run prettier-standard without npx -y.

^{Reviewed by Cursor Bugbot for commit 56abbf2. Bugbot is set up for automated code reviews on this repo. Configure here.}

[coveralls]

coverage: 88.806% (+0.1%) from 88.699% — next into master