Add Neon implementation of find_first_of

[Hari Limaye (hazzlim)]

This PR adds a Neon implementation of find_first_of using the Shuffle approach. For _Pos variants, we select between this and the scalar bitmap approach based on thresholds.

For __std_find_first_not_of_trivial_pos_2, in the case where elements do not fit into the scalar bitmap, it can be better to use the approach of Finding the haystack elements in the needle, so we do this.

Benchmark results:

	MSVC SU	Clang SU
bm/2/3	0.979	0.934
bm/6/81	1	1
bm/7/4	0.949	1.048
bm/9/3	2.804	3.409
bm/22/5	6.136	6.5
bm/58/2	10	11.951
bm/75/85	9.65	10
bm/102/4	11.956	13.677
bm/200/46	9.268	9.5
bm/325/1	1.022	1.017
bm/400/50	9.556	9.783
bm/1011/11	11.201	11.5
bm/1280/46	9.37	9.778
bm/1502/23	9.778	10
bm/2203/54	9.773	10.252
bm/3056/7	13.25	14.261
bm/2/3	0.954	0.978
bm/6/81	1.005	1
bm/7/4	0.979	1.024
bm/9/3	2.02	2.463
bm/22/5	5.286	5.932
bm/58/2	5.462	5.581
bm/75/85	5.669	5.465
bm/102/4	9.333	7.475
bm/200/46	5.281	5.281
bm/325/1	1.098	0.968
bm/400/50	6.603	5.342
bm/1011/11	5.864	6
bm/1280/46	5.745	5.489
bm/1502/23	5.265	5.385
bm/2203/54	5.505	5.5
bm/3056/7	6.667	6.667
bm/2/3	1	0.957
bm/6/81	0.957	1.002
bm/7/4	1.019	0.977
bm/9/3	1.861	2.199
bm/22/5	2.826	2.889
bm/58/2	3.566	3.223
bm/75/85	2.938	2.938
bm/102/4	3.27	3.568
bm/200/46	3	2.897
bm/325/1	1.023	1
bm/400/50	3.009	2.926
bm/1011/11	3.226	3.283
bm/1280/46	3.07	3
bm/1502/23	3.043	3.113
bm/2203/54	2.891	3.017
bm/3056/7	3.302	3.659
bm/2/3	0.904	0.933
bm/6/81	1.062	1
bm/7/4	0.937	0.854
bm/9/3	1.5	1.563
bm/22/5	1.706	1.773
bm/58/2	2.667	1.792
bm/75/85	1.736	1.634
bm/102/4	1.643	1.5
bm/200/46	1.69	1.636
bm/325/1	1.005	0.978
bm/400/50	1.6	1.637
bm/1011/11	1.714	1.643
bm/1280/46	1.606	1.817
bm/1502/23	1.638	1.601
bm/2203/54	1.643	1.669
bm/3056/7	1.627	1.627
bm/2/3	1.042	0.988
bm/6/81	0.978	0.794
bm/7/4	0.977	0.92
bm/9/3	1.174	1.325
bm/22/5	1.244	1.677
bm/58/2	2.58	4.377
bm/75/85	1	1.132
bm/102/4	2.24	3.688
bm/200/46	1	1.333
bm/325/1	5.319	9.287
bm/400/50	1	1.402
bm/1011/11	0.983	1.66
bm/1280/46	1.007	1.533
bm/1502/23	1.012	1.588
bm/2203/54	1	1.564
bm/3056/7	1.674	2.914
bm/2/3	0.872	0.792
bm/6/81	0.96	0.765
bm/7/4	0.975	0.714
bm/9/3	0.966	0.798
bm/22/5	1.308	1.173
bm/58/2	2.15	2.1
bm/75/85	0.991	0.893
bm/102/4	1.979	2.086
bm/200/46	1.028	1.072
bm/325/1	3.932	3.546
bm/400/50	1.029	1.077
bm/1011/11	1.086	1.117
bm/1280/46	0.957	1.108
bm/1502/23	0.976	1.15
bm/2203/54	1.023	1.116
bm/3056/7	1.314	1.366
bm/2/3	0.958	0.709
bm/6/81	1.223	1.127
bm/7/4	2.008	1.484
bm/9/3	2.215	1.077
bm/22/5	4.229	3.394
bm/58/2	7.778	3.035
bm/75/85	1.591	1.515
bm/102/4	7.299	7.219
bm/200/46	1.76	1.741
bm/325/1	10.222	3.258
bm/400/50	1.816	1.736
bm/1011/11	4.2	3.917
bm/1280/46	1.913	1.905
bm/1502/23	2.663	2.506
bm/2203/54	1.87	1.867
bm/3056/7	6.825	6.349
bm/2/3	1.148	1
bm/6/81	1.142	1.024
bm/7/4	0.909	1.029
bm/9/3	0.875	0.979
bm/22/5	0.938	1.074
bm/58/2	1.419	1.893
bm/75/85	1.037	1.217
bm/102/4	0.969	1.218
bm/200/46	0.987	1.256
bm/325/1	2.245	2.904
bm/400/50	1	1.308
bm/1011/11	1.023	1.294
bm/1280/46	0.875	1.278
bm/1502/23	0.891	1.29
bm/2203/54	0.957	1.306
bm/3056/7	1	1.29
bm/2/3	1.016	0.946
bm/6/81	2.16	2.2
bm/7/4	1.646	1.867
bm/9/3	1.436	1.633
bm/22/5	2.442	2.636
bm/58/2	3.389	3.217
bm/75/85	2.878	2.938
bm/102/4	3.173	3.297
bm/200/46	2.809	3.143
bm/325/1	3.115	3.351
bm/400/50	2.943	3.077
bm/1011/11	3.343	3.54
bm/1280/46	3.006	3.07
bm/1502/23	3.182	3.162
bm/2203/54	2.949	3.07
bm/3056/7	3.492	3.81
bm/2/3	1.024	1.011
bm/6/81	0.956	0.782
bm/7/4	1	0.974
bm/9/3	1.272	1.318
bm/22/5	1.524	1.6
bm/58/2	2.833	3.255
bm/75/85	1.059	0.976
bm/102/4	2.864	3.151
bm/200/46	1.173	1.125
bm/325/1	5.581	6.095
bm/400/50	1.152	1.029
bm/1011/11	1.046	1.209
bm/1280/46	1.163	1.117
bm/1502/23	1.136	1.091
bm/2203/54	1.15	1.103
bm/3056/7	1.897	2.316
bm/2/3	1	0.755
bm/6/81	0.968	0.852
bm/7/4	0.976	0.667
bm/9/3	0.933	0.67
bm/22/5	1.193	1.068
bm/58/2	1.967	2.2
bm/75/85	0.953	1
bm/102/4	1.915	2.222
bm/200/46	1	1.217
bm/325/1	3.13	4.286
bm/400/50	0.965	1.205
bm/1011/11	0.987	1.281
bm/1280/46	1	1.222
bm/1502/23	1.02	1.29
bm/2203/54	0.977	1.306
bm/3056/7	1.23	1.602
bm/2/3	0.769	0.73
bm/6/81	0.897	0.752
bm/7/4	1.568	1.17
bm/9/3	1.702	0.764
bm/22/5	2.773	2.439
bm/58/2	5	1.882
bm/75/85	0.856	0.855
bm/102/4	5.535	5.222
bm/200/46	0.839	0.843
bm/325/1	6.147	2.304
bm/400/50	0.832	0.853
bm/1011/11	2.5	2.689
bm/1280/46	0.852	0.869
bm/1502/23	0.979	0.842
bm/2203/54	0.886	0.851
bm/3056/7	4.193	4.372

[Stephan T. Lavavej (StephanTLavavej)]

I'm mirroring this to the MSVC-internal repo. Please notify me if any further changes are pushed, otherwise no action is required.

[Stephan T. Lavavej (StephanTLavavej)]

🕵️ 🦾 🥇