`keepAlive` JSDoc points at `stop()` for a footgun

[janbuchar]

• The keepAlive option says to "use crawler.stop() to exit the crawler gracefully". stop() only halts dispatch; whatever is still queued stays queued. For the case keepAlive is usually reached for — a crawler fed by another crawler or an external producer — calling stop() when the producer finishes drops everything the consumer had not picked up yet (in a local test, 20 of 20).
• Point the JSDoc at taskLoopOptions.isFinishedFunction with (await manager.checkReadiness()).status === 'finished' plus the producer's running flag, and say plainly what stop() does. stop({ drain: true }) would be nice but the doc fix is the important part.