It's silly to have to detect individual CIs. They should identify themselves with the CI environment variable, and most do just that. A manually maintained list of detections will easily get out of date. And if a package using ci-info doesn't update to the latest version all the time, they will not support certain CIs. It also creates unpredictability as you might assume a CI is not supported and then suddenly it gets supported and you didn't account for that. In addition, some of the manual detections are loose and might cause false-positives which could create hard-to-debug bugs.
Why does this even exist? It's just a few lines.
It's not about the number of lines, but rather discoverability and documentation. A lot of people wouldn't even know they need this. Feel free to copy-paste the code if you don't want the dependency. You might also want to read this blog post.