Yeah but there's would be scrappers if the robots file just pointed to a dump file.
Then the scraper could just do a spot check a few dozen random page and check the dump is actually up to date and complete and then they'd know they don't need to waste any time there and move on.
Crawlers are expensive and annoying to run, not to mention unreliable and produce low quality data. If there really were a site dump available, I don't see why it would make sense to crawl the website, except to spot check the dump is actually complete. This used to be standard and it came with open API access for all before the silicon valley royals put the screws on everyone