[Enhancement]: optimize murmurhash64A with SIMD

[Dragonliu2018]

Enhancement
In vectorized operators, such as Hash Group By, there are many scenarios where fixed-length data batches are being hashed in bulk. In such cases, we can use SIMD to optimize.

In particular, the Q36 query in clickbench will benefit from the enhancement.

SELECT  /*+ CB_Q36 parallel(64) */ ClientIP, ClientIP - 1, ClientIP - 2, ClientIP - 3, COUNT(*) AS c FROM hits GROUP BY ClientIP, ClientIP - 1, ClientIP - 2, ClientIP - 3 ORDER BY c DESC LIMIT 10;

[hnwyllmm]

Would you like to take this issue?