I think the provided integration
is extremely easy -- by Intent.
The core scanning library is also extremely easy to integrate; it's just a stand-alone library.
What people sometimes want is a drop-in Activity or something to call. That's not unreasonable, but it's virtually the same experience as calling by Intent.
Some people say they don't want to use a third-party app on the device, usually because they're being paid to build a solution that has this requirement at minimum cost. People who are developing commercial apps with their own better UI don't have this problem, since they don't want to copy the existing UI anyway.
Providing a clone of the UI also means people will start to write in to the OSS project for support since they think they are using BS, and the OSS project isn't for commercial support.
So yes it would only make sense to put in work to support this if people were paying for it, I agree. That's also not unreasonable. I think the license would reasonably be thousands or tens of thousands of dollars depending on volume, putting it out of the range of most contractors being paid to do this cheaply anyway, who would just go copy the source code anyway, license problem or no.
The problem is also: who would get that money? This is the work of lots of people.
So, it's complicated, but my take is that the OSS project solves this problem well for everyone except some contractors, and I don't think the incentives make it reasonable to get the OSS project to do their work even for pay.