Frame Studio and Frame Home: cross-platform builds

Frame Studio now builds for Windows (x64, x86, ARM64), Linux (x64, x86, ARM64)
and macOS (Apple Silicon, Intel):
- studio/platform.h with Win32 and POSIX implementations (dialogs through
  tinyfiledialogs on Linux/macOS), a GLFW window beside the Win32 one, and a
  platform-neutral frame pacer
- MoltenVK support in the renderer (portability enumeration) and bundled in
  the macOS app
- 32-bit builds: large-address-aware, LZMA packing within a 32-bit address space
- unaligned file reads in the loaders go through loadUnaligned (no UB)
- scripts/build-studio.sh, studio/build_windows.bat [x64|x86|arm64],
  cmake/llvm-mingw.cmake, and a GitHub Actions workflow building every target

Frame Home installs without Studio: frame/install-frame-home.{sh,ps1} and
"Install Frame Home.bat" (packaged as FrameHome-installer.zip).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

third_party/astcenc/Source/UnitTest/test_softfloat.cpp

@@ -0,0 +1,68 @@
+// SPDX-License-Identifier: Apache-2.0
+// ----------------------------------------------------------------------------
+// Copyright 2021 Arm Limited
+//
+// Licensed under the Apache License, Version 2.0 (the "License"); you may not
+// use this file except in compliance with the License. You may obtain a copy
+// of the License at:
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS, WITHOUT
+// WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the
+// License for the specific language governing permissions and limitations
+// under the License.
+// ----------------------------------------------------------------------------
+
+/**
+ * @brief Unit tests for the software half-float library.
+ */
+
+#include "gtest/gtest.h"
+
+#include "../astcenc_internal.h"
+
+namespace astcenc
+{
+
+#if (ASTCENC_F16C == 0) && (ASTCENC_NEON == 0)
+
+/** @brief Test normal numbers. */
+TEST(softfloat, FP16NormalNumbers)
+{
+	float result = sf16_to_float((15 << 10) + 1);
+	EXPECT_NEAR(result,  1.00098f, 0.00005f);
+}
+
+/** @brief Test denormal numbers. */
+TEST(softfloat, FP16DenormalNumbers)
+{
+	float result = sf16_to_float((0 << 10) + 1);
+	EXPECT_NEAR(result, 5.96046e-08f, 0.00005f);
+}
+
+/** @brief Test zero. */
+TEST(softfloat, FP16Zero)
+{
+	float result = sf16_to_float(0x0000);
+	EXPECT_EQ(result, 0.0f);
+}
+
+/** @brief Test infinity. */
+TEST(softfloat, FP16Infinity)
+{
+	float result = sf16_to_float((31 << 10) + 0);
+	EXPECT_TRUE(std::isinf(result));
+}
+
+/** @brief Test NaN. */
+TEST(softfloat, FP16NaN)
+{
+	float result = sf16_to_float(0xFFFF);
+	EXPECT_TRUE(std::isnan(result));
+}
+
+#endif
+
+}