Adding pkg-config to Window CI/Wheel builds

.github/workflows/ci-windows.yml

@@ -24,7 +24,7 @@ jobs:
         shell: "bash -l {0}"
         run: >
           conda create --name "viprs_ci" -c conda-forge -c anaconda 
-          python=${{matrix.python-version}} pip wheel compilers openblas -y
+          python=${{matrix.python-version}} pip wheel pkg-config compilers openblas -y
 
       - name: Show info about `viprs_ci` environment
         shell: "bash -l {0}"

.github/workflows/wheels.yml

@@ -20,6 +20,7 @@ jobs:
         os: [ubuntu-latest, windows-latest, macos-13, macos-14]
     env:
       CIBW_SKIP: "pp* cp36-* *-musllinux_* *win32 *_i686 *_s390x"
+      CIBW_BEFORE_BUILD_WINDOWS: "choco install pkgconfiglite"
 
     steps:
       - uses: actions/checkout@v4
@@ -53,6 +54,7 @@ jobs:
         run: | 
           python -m pip install --upgrade pip
           python -m pip install dist/viprs*.tar.gz
+          python -m viprs_fit -h
 
       - uses: actions/upload-artifact@v4
         with:

docs/installation.md

@@ -11,6 +11,7 @@ Building the `viprs` package requires the following dependencies:
 * `C/C++` Compilers
 * `cython`
 * `numPy` 
+* `pkg-config`
 * `sciPy` (>=1.5.4)
 
 To take full advantage of the **parallel processing** capabilities of the package, you will also need to make sure that 
@@ -21,7 +22,7 @@ the following packages/libraries are available:
 
 ### Setting up the environment with `conda`
 
-If you can use `Anaconda` or `miniconda` to manage your Python environment, we recommend using them to create 
+If you can use `Anaconda` or `miniconda` to manage your Python environment, we **recommend** using them to create 
 a new environment with the required dependencies as follows:
 
 ```bash