SIMD Conversion Functions

[RalphSteinhagen]

This PR introduces a set of conversion functions that for production use should be inlined/compile-time-merged.
These functions leverage SIMD capabilities to improve the performance of complex data transformations. Original vectorisiation example by @mattkretz: https://godbolt.org/z/7MT71MbGz

New Conversion Blocks:

• ComplexToInterleaved: Converts a stream of complex numbers to an interleaved stream of real and imaginary components.
• InterleavedToComplex: Converts an interleaved stream of real and imaginary components back to a stream of complex numbers.
• Abs: Computes the magnitude of complex or arithmetic input streams.
• Real: Extracts the real component of complex input streams.
• Imag: Extracts the imaginary component of complex input streams.
• Arg: Computes the argument (phase angle) of complex input streams.
• RadiansToDegree: Converts radians to degrees.
• DegreeToRadians: Converts degrees to radians.
• ToRealImag: Decomposes complex numbers into their real and imaginary components.
• RealImagToComplex: Combines real and imaginary components into complex numbers.
• ToMagPhase: Decomposes complex numbers into their magnitude and phase components.
• MagPhaseToComplex: Combines magnitude and phase components into complex numbers.

Performance Assessment:

The following table summarizes the nominal clock cycles for each assembly instruction, demonstrating the efficiency improvements achieved through SIMD vectorization. Some functions do not (yet) use SIMD implementation due to the missing SIMD-permute API. This should be improved once this API becomes available.

As an illustrative example to assess the asm-related efficiency of various possible implementations: https://godbolt.org/z/obexWhjMq

Function	x86_64 (Zen3)		ARM (Raspberry Pi 5)
	Latency	Throughput	Latency	Throughput
ProcessOne	4 cycles	1 cycle	4 cycles	1 cycle
ProcessOneScalling	10 cycles	3 cycles	9 cycles	4 cycles
ProcessOneClamp	16 cycles	8 cycles	12 cycles	4 cycles
complexToComponents	8 cycles	4 cycles	6 cycles	3 cycles
complexToArg	60 cycles	15 cycles	80 cycles	20 cycles
complexToArg2	60 cycles	15 cycles	80 cycles	20 cycles
complexToArgGeneric	45 cycles	11 cycles	60 cycles	15 cycles
complexToArgGeneric2	15 cycles	3 cycles	5 cycles	2 cycles
complexToAbsGeneric	45 cycles	11 cycles	60 cycles	15 cycles
complexToAbsGeneric2	10 cycles	2 cycles	3 cycles	1 cycle
complexToAbs	12 cycles	6 cycles	14 cycles	7 cycles
complexToAbs2	12 cycles	6 cycles	14 cycles	7 cycles

[drslebedev]

Thank you for your contribution!

Overall, it looks good, and the unit tests are well done.

I’ve suggested several changes in the comments regarding the compilation error for clang and conversion warnings.

Please review the comments, once these issues are resolved, the PR can be merged.

[drslebedev]

There are several issues with the failing test including: 1) qa_Message, 2) qa_FilterTool, 3) long compilation times for gcc, 4) during the first run, the qa_DataSink for emcc | Debug test also failed due to timeouts, but it passed upon rerun.

This PR will be merged, but the remaining issues and the problems with the flaky tests need to be addressed and investigated further.