feat: add snmalloc flags as options to SCQ

[gangliao]

Issue #, if available:

apache/arrow#9206

Description of changes:

The performance improvement is huge that can help us to beat any Java-related OLAP framework over batches and streams.

w/o these features

Query 12 iteration 0 took 241.4 ms
Query 12 iteration 1 took 234.8 ms
Query 12 iteration 2 took 229.8 ms
Query 12 iteration 3 took 229.5 ms
Query 12 iteration 4 took 228.3 ms
Query 12 iteration 5 took 230.0 ms
Query 12 iteration 6 took 228.3 ms
Query 12 iteration 7 took 229.3 ms
Query 12 iteration 8 took 229.9 ms
Query 12 iteration 9 took 230.1 ms
Query 12 avg time: 231.13 ms

with lto + simd + snmalloc:

Query 12 iteration 0 took 118.8 ms
Query 12 iteration 1 took 118.7 ms
Query 12 iteration 2 took 118.3 ms
Query 12 iteration 3 took 117.0 ms
Query 12 iteration 4 took 117.7 ms
Query 12 iteration 5 took 118.9 ms
Query 12 iteration 6 took 117.9 ms
Query 12 iteration 7 took 118.1 ms
Query 12 iteration 8 took 117.7 ms
Query 12 iteration 9 took 117.8 ms
Query 12 avg time: 118.09 ms

By submitting this pull request

• I confirm that my contribution is made under the terms of the Apache 2.0 license.
• I confirm that I've made a best effort attempt to update all relevant documentation.