rv64v: fast path for unit-stride ld/st; vectorize this fast path

[shinezyy]

• Remove useless parameters for get_mask, which is misleading
• For unit-stride ld/st, perform a dummy memory access to obtain host addr. If host TLB miss or crossing page, then we fall back to the slow path. This avoids redundant address translation.
• Implement masked ld/st with bit operations instread of branches, which makes code easier to vectorize
• Always prefer bounded loop to make it easier to vectorize
• Add optimization flags: -flto -ftree-vectorize -march=native
• Allow to check fast vse with slow path and store commit difftest
• This patch improves the simualtion speed of vectorized h264_sss by 3x, improves the simualtion speed of early 2G instructions of vectorized h264_sss by 9x.

[shinezyy]