Add SVE2 Bgr48pToBgra32 optimization

[ermig1979]

Summary

• Add SVE2 implementation of Bgr48pToBgra32 using scalable predicates for body and tail processing.
• Dispatch the public API to the SVE2 path when available and add SVE2 auto-test coverage.
• Add the new source to the VS2022 SVE2 project and document it in release 7.2.163.

Validation

• aarch64-linux-gnu-g++ -march=armv9-a+sve2 -msve-vector-bits=scalable -std=c++11 -O3 -fPIC -I src -c src/Simd/SimdSve2BgrToBgra.cpp -o /tmp/SimdSve2BgrToBgra.o
• aarch64-linux-gnu-g++ -march=armv9-a+sve2 -msve-vector-bits=scalable -std=c++11 -O3 -fPIC -I src -c src/Simd/SimdLib.cpp -o /tmp/SimdLib-sve2.o
• cmake --build build --parallel$(nproc)
• export LD_LIBRARY_PATH="$(pwd):$LD_LIBRARY_PATH" && ./Test "-r=.." -fi=Bgr48pToBgra32 -tt=1 -ts=1