Add an option to only apply limit to specific user agents

I worry this is a very brittle approach. I've spent a lot of time staring at access logs over the last year, and I've seen a big issue is that these bot requests are coming from many different sources, with requests that do include known bots, but seemingly more often from what some WAF services call "Browser impersonators", which can be a combination of any versions of OS and Browser. That's just going to be too diverse to capture with a straight string matching comparison.

I might suggest checking string position, or a regex comparison instead? There is also the Crawler Rate Limit module, which can rate limit bots, outside of a faceted search context.