[X86] Cast to __v4hi instead of __m64 in the implementation of...

[X86] Cast to __v4hi instead of __m64 in the implementation of _mm_extract_pi16 and _mm_insert_pi16.

__m64 is a vector of 1 long long. But the builtins these intrinsics
are calling expect a vector of 4 shorts.

Fixes PR44589