Fix endianness detection for MSVC on ARM/ARM64 [skipp appveyor]

As reported by #287

Author: tbeu

.github/workflows/cmake.yml

@@ -104,6 +104,26 @@ jobs:
       - name: Test
         run: ctest --no-tests=error --test-dir build --build-config ${{ matrix.configuration }} --parallel 8
 
+  build-windows-arm:
+    name: windows-arm-msvc
+    runs-on: windows-11-arm
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v6
+        with:
+          fetch-depth: 5
+          submodules: recursive
+      - name: Setup python environment
+        uses: actions/setup-python@v6
+        with:
+          python-version: 3.13
+      - name: Configure
+        run: cmake -S . -B build -A ARM64 -DMATIO_SHARED=OFF -DMATIO_MAT73=OFF -DMATIO_WITH_ZLIB=OFF
+      - name: Build with msvc
+        run: cmake --build build --config Release
+      - name: Test
+        run: ctest --no-tests=error --test-dir build --build-config Release --parallel 8
+
   build-openbsd:
     name: openbsd-clang
     runs-on: ubuntu-latest

src/matio_private.h

@@ -90,10 +90,11 @@
     defined(__ppc__) || defined(__hpux) || defined(_MIPSEB) || defined(_POWER) ||          \
     defined(__s390__)
 #define MATIO_BE
-#elif defined(__i386__) || defined(__alpha__) || defined(__ia64) || defined(__ia64__) ||   \
-    defined(_M_IX86) || defined(_M_IA64) || defined(_M_ALPHA) || defined(__amd64) ||       \
-    defined(__amd64__) || defined(_M_AMD64) || defined(__x86_64) || defined(__x86_64__) || \
-    defined(_M_X64) || defined(__bfin__) || defined(__loongarch64) || defined(__aarch64__)
+#elif defined(__i386__) || defined(__alpha__) || defined(__ia64) || defined(__ia64__) ||      \
+    defined(_M_IX86) || defined(_M_IA64) || defined(_M_ALPHA) || defined(__amd64) ||          \
+    defined(__amd64__) || defined(_M_AMD64) || defined(__x86_64) || defined(__x86_64__) ||    \
+    defined(_M_X64) || defined(__bfin__) || defined(__loongarch64) || defined(__aarch64__) || \
+    defined(_M_ARM64) || defined(_M_ARM)
 #define MATIO_LE
 #endif