Fix pinnings for ppc64le with enabled CUDA

[mpigou]

Following the merges of #7660 and #7653, the rerender of my recipe fails with:

ValueError: Variant configuration errors in /home/fs/.cache/conda-smithy/conda_build_config.yaml:
  zip fields in zip_key group frozenset(
    {'cuda_compiler_version',
     'c_stdlib_version',
     'cxx_compiler_version',
     'fortran_compiler_version',
     'c_compiler_version'}
  ) are not all the same length

This is due to having two output packages, one without CUDA support but built for all platforms including linux_ppc64le, and one with CUDA support but restricted to linux_64 and linux_aarch64. In this edge case, for the linux_ppc64le platform, the resulting pinnings are:

c_compiler_version:            # [unix]
  - 14                         # [linux]
c_stdlib_version:              # [unix]
  - 2.17                       # [linux]
  - 2.17                       # [linux and os.environ.get("CF_CUDA_ENABLED", "False") == "True"]
cxx_compiler_version:          # [unix]
  - 14                         # [linux]
fortran_compiler_version:      # [unix or win64]
  - 14                         # [linux]
cuda_compiler_version:
  - None

I thus propose using the same selector on c_stdlib_version than on other keys to remove the second entry for the ppc64le platform:

c_stdlib_version:              # [unix]
  - 2.17                       # [linux]
  - 2.17                       # [linux and not ppc64le and os.environ.get("CF_CUDA_ENABLED", "False") == "True"]
  - 10.13                      # [osx and x86_64]
  - 11.0                       # [osx and arm64]

[conda-forge-admin]

Hi! This is the friendly automated conda-forge-linting service.

I just wanted to let you know that I linted all conda-recipes in your PR (recipe/meta.yaml) and found it was in an excellent condition.

[h-vetinari]

Thank you very much. Indeed, that was due to those two PRs crossing wires.

[h-vetinari]

I also double-checked that we don't need the same selector modification here:

conda-forge-pinning-feedstock/recipe/conda_build_config.yaml

Line 153 in ed573fb

- c_stdlib_version # [linux and os.environ.get("CF_CUDA_ENABLED", "False") == "True"]

This is fine because the cuda 11.8 migrator sets c_stdlib_version unconditionally.