[WIP] Initial SIMD support

[andreaTP]

I wanted to relax a bit, and took this yak shaving task.
I realized that the SIMD support needs a few little tweaks to the public API, and, the initial support, requires quite a bit of wiring rising the bar for possible contributions.

Notes:

• I discussed with @evacchi and the usage of the Stack with double push and double pop comes straight from wazero I think it's a good approach for further optimizations!
• the vector API, at the moment requires copy of arrays, I should resurrect and finalize Exploration - long array primitive MStack #447 to improve here
• there is only one operation actually implemented as a Vector instruction V128_NOT but demonstrates that the entire wiring works
• I'll clean up the PR in the next days and try to finalize it (with CI setup etc.)

[evacchi]

I am not 100% sold on the approach Map<Opcode, OpImpl>, especially because OpImpl causes the indirection of a method reference. All SIMD opcodes are prefixed with 0xFD so one approach could be

• in InterpreterMachine, add a new case with SIMD_PREFIX(0xFD);
• if SIMD support is enabled, then delegate to another class/method that implements the switch table for that set of opcodes

while this requires adding SIMD_PREFIX(0xFD) to the Opcode enum, it also allows to raise a proper error message if the feature is not enabled.

[andreaTP]

Thanks for your proposal @evacchi , I have no strong opinions here, happy to go down with it if there is agreement, let's @bhelx or @electrum break the tie 🙂