[CORE] Avoid copy in ByteLiteralNode

[jinchengchenghh]

The serialized bloom filter is binary node, it may be very large, then affect the performance.

[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

[github-actions]

Run Gluten Clickhouse CI

[github-actions]

Run Gluten Clickhouse CI

[github-actions]

Run Gluten Clickhouse CI

[jinchengchenghh]

Can not use ByteBuffer to serialize substrait plan because of exception

java.io.NotSerializableException: java.nio.DirectByteBuffer

[jinchengchenghh]

/Benchmark Velox

[jinchengchenghh]

/Benchmark Velox TPCDS

[GlutenPerfBot]

===== Performance report for TPCDS SF2000 with Velox backend, for reference only ====

query	log/native_5763_time.csv	log/native_master_05_15_2024_2984bd740_time.csv	difference	percentage
q1	15.01	18.13	3.123	120.80%
q2	18.51	14.70	-3.808	79.42%
q3	4.37	4.30	-0.064	98.54%
q4	62.88	61.79	-1.096	98.26%
q5	8.09	7.96	-0.127	98.43%
q6	3.97	3.75	-0.225	94.32%
q7	4.25	5.67	1.412	133.18%
q8	4.98	3.77	-1.211	75.70%
q9	16.32	18.07	1.747	110.70%
q10	9.08	9.46	0.377	104.15%
q11	36.06	34.82	-1.236	96.57%
q12	3.13	2.40	-0.730	76.66%
q13	5.83	5.58	-0.253	95.67%
q14a	42.08	44.17	2.092	104.97%
q14b	43.64	43.23	-0.414	99.05%
q15	2.44	2.47	0.037	101.53%
q16	38.94	39.90	0.954	102.45%
q17	4.67	7.15	2.483	153.21%
q18	5.78	5.97	0.193	103.34%
q19	3.20	2.18	-1.020	68.17%
q20	3.87	2.68	-1.188	69.30%
q21	1.13	1.00	-0.133	88.22%
q22	7.91	7.87	-0.041	99.49%
q23a	81.84	81.08	-0.767	99.06%
q23b	102.37	99.94	-2.433	97.62%
q24a	80.55	77.77	-2.776	96.55%
q24b	78.65	69.48	-9.171	88.34%
q25	9.98	4.31	-5.665	43.23%
q26	2.82	2.63	-0.184	93.48%
q27	2.70	4.71	2.006	174.19%
q28	22.49	21.78	-0.706	96.86%
q29	8.32	8.32	0.001	100.01%
q30	4.16	5.58	1.419	134.13%
q31	7.09	5.78	-1.312	81.50%
q32	1.00	1.10	0.096	109.57%
q33	5.73	5.73	-0.004	99.93%
q34	4.97	5.46	0.489	109.85%
q35	6.49	6.64	0.149	102.30%
q36	3.13	3.11	-0.014	99.56%
q37	3.91	3.84	-0.069	98.23%
q38	11.85	11.79	-0.051	99.57%
q39a	3.24	3.27	0.030	100.92%
q39b	3.13	3.29	0.164	105.25%
q40	3.56	3.61	0.052	101.46%
q41	0.59	0.55	-0.048	91.85%
q42	0.81	0.83	0.014	101.78%
q43	3.49	3.59	0.093	102.65%
q44	6.38	7.10	0.722	111.32%
q45	3.28	3.53	0.250	107.61%
q46	3.02	3.11	0.086	102.84%
q47	14.21	17.94	3.729	126.25%
q48	4.39	4.84	0.443	110.09%
q49	7.21	7.39	0.180	102.49%
q50	22.36	30.92	8.563	138.30%
q51	8.46	8.59	0.138	101.63%
q52	0.94	1.12	0.181	119.14%
q53	1.62	1.75	0.123	107.56%
q54	3.18	3.11	-0.066	97.93%
q55	0.92	1.04	0.114	112.41%
q56	5.39	5.35	-0.038	99.29%
q57	8.41	8.49	0.080	100.96%
q58	2.48	2.39	-0.089	96.41%
q59	14.17	15.43	1.259	108.88%
q60	8.43	5.87	-2.561	69.62%
q61	6.11	8.72	2.608	142.66%
q62	3.59	3.90	0.306	108.53%
q63	2.99	1.93	-1.061	64.53%
q64	50.12	48.79	-1.338	97.33%
q65	13.67	13.50	-0.167	98.78%
q66	2.81	2.95	0.133	104.74%
q67	358.17	350.79	-7.382	97.94%
q68	6.56	4.02	-2.547	61.19%
q69	10.27	7.71	-2.551	75.15%
q70	7.45	8.44	0.983	113.19%
q71	2.21	4.29	2.084	194.46%
q72	190.82	189.56	-1.262	99.34%
q73	2.19	2.15	-0.041	98.14%
q74	20.97	20.93	-0.042	99.80%
q75	24.74	25.12	0.371	101.50%
q76	7.14	8.64	1.498	120.99%
q77	1.68	1.76	0.080	104.78%
q78	38.38	38.03	-0.352	99.08%
q79	3.32	3.47	0.155	104.67%
q80	11.51	11.54	0.025	100.22%
q81	5.10	5.67	0.563	111.03%
q82	6.23	6.27	0.041	100.65%
q83	1.47	1.56	0.098	106.66%
q84	3.10	2.63	-0.473	84.73%
q85	7.34	7.92	0.584	107.95%
q86	3.21	2.95	-0.260	91.90%
q87	12.12	13.32	1.205	109.94%
q88	16.80	16.55	-0.256	98.48%
q89	4.26	2.74	-1.518	64.33%
q90	6.66	6.45	-0.214	96.79%
q91	2.63	2.60	-0.036	98.65%
q92	1.12	1.26	0.142	112.75%
q93	28.65	28.79	0.135	100.47%
q94	21.29	21.15	-0.132	99.38%
q9	87.06	86.57	-0.494	99.43%
q5	3.01	2.51	-0.507	83.19%
q96	12.26	12.37	0.117	100.95%
q97	1.88	1.86	-0.027	98.58%
q98	8.25	8.93	0.683	108.27%
q99	8.25	8.93	0.683	108.27%
total	1911.03	1897.48	-13.549	99.29%

[jinchengchenghh]

/Benchmark Velox