Document Google-Extended and Applebot-Extended

[blaine-arcjet]

It seems that Google and Apple use "Extended" as signals in a robot.txt to exclude the scraped data from their LLMs. Since this is not sent as part of a User-Agent header, we can't block it via our bot detection. Instead, we should recommend users set up their robot.txt with these signals.

Ref arcjet/well-known-bots#13
Ref arcjet/well-known-bots#14