New MINOR release 03.27.00

Changes since 03.26.00:

- fix: RawConverter string conversion toRaw
- fix(SubdeviceBackend): Doxygen warnings and doxygen formatting
- chore: remove obsolete IEEE754_SingleConverter and FixedPointConverter
- chore: add IEEE754 tests for RawConverter
- feat(SubDevice): use RawConverter instead of FixedPointConverter
- fix: 64 bits in RawConverter
- feat(SubDeviceBackend): register window interface
- fix: coding style in SubdeviceBackend
- fix: callForRawType calls with 32 bits instead of 64 bits
- fix: remove unneeded includes
- chore: replace FixedPointConverter with RawConverter in bit accessor
- fix: add missing int64 bit raw type in callForRawType
- chore: remove unused source files
- chore: replace FixedPointConverter with RawConverter in DummyAccessors
- fix: compiler/linter warning
- fix: compiler error with clang
- fix: unaligned data access is UB
- fix(test): cast out-of-range float to uint8 is UB
- fix(test): casting value outside target range is UB
- feat/chore: added std::tuple replacements for some boost MPL things
- fix: directly include headers, not indirectly
- chore: performance optimisations for muxed accessor (cf. #15434)
- fix: insufficient ChannelInfo comparison operator
- chore: performance optimisations for the RawConverter (cf. #15434)
- fix: potentially unclean ShmDummy termination
- feat: introduce RawConverter to replace FixedPointConverter
- fix: raw type in channel of numeric catalogue
- fixup: correct addresses in testjsonparser correct address string in jsonparser.
- fix: code formatting
- feat: Implement double buffer pushtype (part 1) - Extend NumericAddressedRegisterCatalogue - Extend JsonMapFileParser - Introduce DoubleBufferAccessor.
- fix: MatchingMode=historized might use initial values
- chore: replace boost::numeric::converter
- feat: add numeric converter
- chore: remove option to throw boost::numeric::bad_numeric_cast
- chore: move Boolean and Void to separate headers
- fix: wrong include of own header