[ned14-internal-quickcpplib] Update port @2022-07-04

[BurningEnlightenment]

Describe the pull request

• What does your PR fix?

  Updates quickcpplib, integrates the upstream dependency switch from gsl-lite to span-lite.

  Additionally I applied the feedback wrt polyfill features from [stduuid] Add new port @1.2.2 #25115

• Which triplets are supported/not supported? Have you updated the CI baseline?

  Set the Windows arm64 baseline to fail, because it is caused by an upstream bug algorithm/hash is incompatible with MSVC arm64 ned14/quickcpplib#32

• Does your PR follow the maintainer guide?

  Yes

• If you have added/updated a port: Have you run ./vcpkg x-add-version --all and committed the result?

  Yes

[BurningEnlightenment]

I can't reproduce the linux build failure locally and I don't have access to an windows arm64 device. So I need access to the (cmake) build logs.

[dg0yt]

I can't reproduce the linux build failure locally and I don't have access to an windows arm64 device. So I need access to the (cmake) build logs.

You can click through from Checks to Azure pipelines to artifacts, and download the failure logs. The menu for download appears behind the three dots on the right side of an artifact, when the mouse is over that line.

[BurningEnlightenment]

Ah, I missed the artifacts link, thank you.

Ok, so the gcc version on linux is too old for std::span and would need the polyfill-cxx20 feature. Is there a way to enable this just for the CI?

I will set the Windows arm64 baseline to fail, because it is caused by an upstream bug ned14/quickcpplib#32

[dg0yt]

Ok, so the gcc version on linux is too old for std::span and would need the polyfill-cxx20 feature. Is there a way to enable this just for the CI?

As an idea, for feedback from the core maintainers: A CI-only activation could be achieved via a test port (e.g. a new empty vcpkg-ci-polyfill port) in scripts/test-ports which depends on quickcpplib[polyfill-cxx20] for linux.

[BurningEnlightenment]

@Adela0814 is this sufficient?