Add SIMD operations that use f16 and f128

Eventually we will want to be able to make use of simd operations for f16 and f128, now that we have primitives to represent them. Possibilities that I know of:

• Aarch64 neon supports float16x{4,8} https://developer.arm.com/architectures/instruction-sets/intrinsics/#f:@navigationhierarchiessimdisa=[Neon]&f:@navigationhierarchiesreturnbasetype=[float]&f:@navigationhierarchieselementbitsize=[16]&q=.
  • Implementation: https://github.com/rust-lang/stdarch/pull/1726
  • Tracking issue: https://github.com/rust-lang/rust/issues/136306
• Arm sve supports float16x{1,2} https://developer.arm.com/architectures/instruction-sets/intrinsics/#f:@navigationhierarchiesreturnbasetype=[float]&f:@navigationhierarchieselementbitsize=[16]&f:@navigationhierarchiessimdisa=[sve2,sve]&q=
  • We probably can't support this until our runtime-sized types story is better
• RISC-V apparently has both f16 and f128 https://five-embeddev.com/riscv-user-isa-manual/riscv-user-2.2/v.html
  • Similarly relies on the "V" extension which has runtime-sized types https://github.com/rust-lang/rust/issues/114544
• NVIDIA PTX has f16 SIMD
  • Implementation: https://github.com/rust-lang/stdarch/pull/1626
  • Submodule https://github.com/rust-lang/rust/pull/128866
  • Tracking issue: https://github.com/rust-lang/rust/issues/111199
• x86 with +avx512fp16
  • Implementation: https://github.com/rust-lang/stdarch/pull/1605
  • Submodule: https://github.com/rust-lang/rust/pull/128466
  • Tracking issue: https://github.com/rust-lang/rust/issues/127213
• Portable SIMD should eventually be able to support these operations

Probably some work/research overlap with adding assembly https://github.com/rust-lang/rust/issues/125398

Tracking issue: https://github.com/rust-lang/rust/issues/116909

Start with tracking issue #116909 and review the linked stdarch and Rust submodule pull requests for AArch64, NVIDIA PTX, and x86. Compare those approaches with the runtime-sized type constraints and the related assembly issue #125398; the work is complete when a specific target scope and implementation path for f16 or f128 SIMD operations is established and implemented.