Add vmull_p64 and vmull_high_p64 for aarch64

[SparrowLii]

FIXES #1156
This PR serves as a supplement to #1147
Although vmull_p64 is temporarily unavailable on Arm32 due to the llvm‘s unsupporting, we can still make it available on aarch64.

[rust-highfive]

r? @Amanieu

(rust-highfive has picked a reviewer for you, use r? to override)

[LuoZijun]

Why they have different type signatures ？

stdarch:

pub type p64  = u64;
pub type p128 = u128;

pub unsafe fn vmull_p64(a: p64, b: p64) -> p128 {
     ...
}

AARCH64 Neon Intrinsics: https://developer.arm.com/architectures/instruction-sets/simd-isas/neon/intrinsics?search=vmull_p64

poly128_t  vmull_p64 (poly64_t a, poly64_t b)
poly128_t  vmull_high_p64 (poly64x2_t a, poly64x2_t b)

[SparrowLii]

This is to be consistent with other primitive variables(u64, s64). float64_t will also be replaced with f64 in function signatures.
You can look at https://github.com/rust-lang/stdarch/blob/master/crates/stdarch-verify/src/lib.rs#L252 and https://github.com/rust-lang/stdarch/blob/master/crates/stdarch-verify/tests/arm.rs#L824. Both p64 and poly64_t are regarded as P64 type.