Correct CUDA linking with big thanks to eagle-eyed @rcurtin (closes #6)

[eddelbuettel]

As discussed in #6 and with a heartfelt Thank You!! to @rcurtin

[Copilot]

Pull Request Overview

This PR corrects CUDA linking configuration by adding the missing -lcuda (CUDA Driver API library) to the linker flags. This fix was identified and reported by @rcurtin in issue #6.

• Adds -lcuda to the CUDA_LIBS variable in both configure.ac and the generated configure script

Reviewed Changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated no comments.

File	Description
configure.ac	Added -lcuda to CUDA_LIBS variable to link against CUDA Driver API library
configure	Regenerated configure script reflecting the updated CUDA_LIBS from configure.ac

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[rcurtin]

Happy to help out!

[eddelbuettel]

I blame cmake, as usual, which I normally do not run in verbose mode so I did not see the linker lines. And of course I didn't think of that obvious a library missing -- plus the linker errors were all clSomething. Oh well. Blame still on me 😀

[coatless]

@eddelbuettel thanks for the PR and effort. I'm sorry for missing the most important flag for CUDA. 😭

@rcurtin thanks for the second look!