Implement dynamic recompilation or JIT

[TrueGoric]

Will be needed earlier than I thought, the interpreted execution is awfully slow (4.3-4.5 MHz on NOPs, optimized assembly) and JIT is reported to be very efficient.

See:

• https://www.reddit.com/r/EmuDev/comments/fxrcf1/some_benchmarks_with_dynamic_recompilation_in_c/
• https://www.reddit.com/r/EmuDev/comments/g1bnn6/another_view_on_some_benchmarks_with_dynamic/
• https://github.com/exelix11/Chip8CIL