[VL] Sort shuffle writer use vectorized c2r

[marin-ma]

	Default	with PR
time of input iterator	397.95	403.29
time of scan and filter	246.58	253.03
not_counted	216.78	224.2
time to shuffle c2r	159.02	134.57
time to compress	93.77	93.09
time to deserialize	67.05	67.93
time to decompress	49.18	49.85
shuffle write time	46.83	41.75
time of rowConstruction	43.07	44.19
time to shuffle sort	42.95	42.99
time of project	42.16	41.24
time of hash probe	41.04	40.54
idle	34.21	23.12
time of aggregation	14.53	14.33
time of postProjection	7.83	7.83

[github-actions]

Thanks for opening a pull request!

Could you open an issue for this pull request on Github Issues?

https://github.com/apache/incubator-gluten/issues

Then could you also rename commit message and pull request title in the following format?

[GLUTEN-${ISSUES_ID}][COMPONENT]feat/fix: ${detailed message}

See also:

• Other pull requests

[marin-ma]

TPCH SF3T perf, totaltime shuffle c2r 162s -> 138s

query	log/native_6782_time.csv	log/native_master_08_12_2024_time.csv	difference	percentage
q1	25.08	25.41	0.329	101.31%
q2	32.22	32.31	0.086	100.27%
q3	47.38	48.97	1.589	103.35%
q4	37.90	37.56	-0.333	99.12%
q5	130.85	129.34	-1.513	98.84%
q6	9.50	11.71	2.211	123.27%
q7	115.07	112.57	-2.506	97.82%
q8	117.24	118.42	1.177	101.00%
q9	158.84	163.42	4.584	102.89%
q10	55.32	58.64	3.314	105.99%
q11	29.84	29.64	-0.198	99.34%
q12	24.39	25.43	1.037	104.25%
q13	32.53	33.65	1.125	103.46%
q14	15.10	16.73	1.633	110.82%
q15	30.28	33.66	3.378	111.15%
q16	18.35	17.96	-0.393	97.86%
q17	136.81	138.44	1.633	101.19%
q18	168.52	169.66	1.146	100.68%
q19	19.59	16.46	-3.129	84.03%
q20	31.96	36.92	4.960	115.52%
q21	263.55	268.29	4.745	101.80%
q22	18.27	18.18	-0.089	99.51%
total	1518.60	1543.38	24.785	101.63%

[github-actions]

This PR is stale because it has been open 45 days with no activity. Remove stale label or comment or this will be closed in 10 days.

[marin-ma]

@

TPCH SF3T perf, totaltime shuffle c2r 162s -> 138s

query log/native_6782_time.csv log/native_master_08_12_2024_time.csv difference percentage
q1 25.08 25.41 0.329 101.31%
q2 32.22 32.31 0.086 100.27%
q3 47.38 48.97 1.589 103.35%
q4 37.90 37.56 -0.333 99.12%
q5 130.85 129.34 -1.513 98.84%
q6 9.50 11.71 2.211 123.27%
q7 115.07 112.57 -2.506 97.82%
q8 117.24 118.42 1.177 101.00%
q9 158.84 163.42 4.584 102.89%
q10 55.32 58.64 3.314 105.99%
q11 29.84 29.64 -0.198 99.34%
q12 24.39 25.43 1.037 104.25%
q13 32.53 33.65 1.125 103.46%
q14 15.10 16.73 1.633 110.82%
q15 30.28 33.66 3.378 111.15%
q16 18.35 17.96 -0.393 97.86%
q17 136.81 138.44 1.633 101.19%
q18 168.52 169.66 1.146 100.68%
q19 19.59 16.46 -3.129 84.03%
q20 31.96 36.92 4.960 115.52%
q21 263.55 268.29 4.745 101.80%
q22 18.27 18.18 -0.089 99.51%
total 1518.60 1543.38 24.785 101.63%

@FelixYBW We need to merge this patch to enable the c2r optimization in shuffle. This performance data was measured a few months ago. I will refresh it when jenkins server is back.