URLs in, their content out; no project.

One url waits for the page and returns it -- up to timeout seconds (60 by default, 120 at most), then 202 with the id to poll, the fetch carrying on either way. A list of urls is always 202: grouped by host, small groups fanned out a task per URL, and either polled at GET /scrape/{id} or announced by webhook_url (scrape.completed, signed with the secret returned here once).