Add net8.0 target and AVX-512/Vector512 parsers

[nietras]

• Use nuint for masks and only use Vector512 in 64-bit process.
• Add benchmark results to benchmarks directory.
• .NET 8.0 Preview 7 appears to offer nice minor improvements across everything. Some major for floats.

[codecov-commenter]

Codecov Report

Patch coverage is 100.00% of modified lines.

Files Changed	Coverage
src/Sep/Internals/SepArrayExtensions.cs	ø
src/Sep/Internals/SepToStringHashPool.cs	ø
src/Sep/Internals/SepParseMask.cs	100.00%
...p/Internals/SepParserAvx2PackCmpOrMoveMaskTzcnt.cs	100.00%
...Internals/SepParserAvx512PackCmpOrMoveMaskTzcnt.cs	100.00%
src/Sep/Internals/SepParserFactory.cs	100.00%
src/Sep/Internals/SepParserIndexOfAny.cs	100.00%
...p/Internals/SepParserSse2PackCmpOrMoveMaskTzcnt.cs	100.00%
...p/Internals/SepParserVector128NrwCmpExtMsbTzcnt.cs	100.00%
...p/Internals/SepParserVector256NrwCmpExtMsbTzcnt.cs	100.00%
... and 6 more

📢 Thoughts on this report? Let us know!.

[nietras]

BenchmarkDotNet=v0.13.5, OS=Windows 10 (10.0.17763.3287/1809/October2018Update/Redstone5)
Intel Xeon Silver 4316 CPU 2.30GHz, 1 CPU, 40 logical and 20 physical cores
.NET SDK=8.0.100-preview.7.23376.3
  [Host]     : .NET 8.0.0 (8.0.23.37506), X64 RyuJIT AVX2
  Job-TNYKAJ : .NET 8.0.0 (8.0.23.37506), X64 RyuJIT AVX2

Runtime=.NET 8.0  Toolchain=net80  IterationTime=300.0000 ms  
MaxIterationCount=9  MinIterationCount=3  WarmupCount=5  

Method	Parser	Filler	Mean	Error	StdDev	Ratio	MB/s	Allocated	Alloc Ratio
Parse	Avx2PackCmpOrMoveMaskTzcnt	0_Assets___ 00000279	49.65 ns	0.636 ns	0.165 ns	1.00	10717.9	-	NA
Parse	Avx512PackCmpOrMoveMaskTzcnt	0_Assets___ 00000279	46.77 ns	0.801 ns	0.208 ns	1.00	11378.0	-	NA
Parse	IndexOfAny	0_Assets___ 00000279	387.26 ns	7.467 ns	1.939 ns	1.00	1374.1	-	NA
Parse	Sse2PackCmpOrMoveMaskTzcnt	0_Assets___ 00000279	57.37 ns	0.844 ns	0.046 ns	1.00	9276.3	-	NA
Parse	Vector128NrwCmpExtMsbTzcnt	0_Assets___ 00000279	71.10 ns	1.258 ns	0.449 ns	1.00	7485.0	-	NA
Parse	Vector256NrwCmpExtMsbTzcnt	0_Assets___ 00000279	51.60 ns	0.864 ns	0.308 ns	1.00	10312.0	-	NA
Parse	Vector512NrwCmpExtMsbTzcnt	0_Assets___ 00000279	48.39 ns	0.578 ns	0.150 ns	1.00	10996.9	-	NA

BenchmarkDotNet=v0.13.5, OS=Windows 10 (10.0.17763.3287/1809/October2018Update/Redstone5)
Intel Xeon Silver 4316 CPU 2.30GHz, 1 CPU, 40 logical and 20 physical cores
.NET SDK=8.0.100-preview.7.23376.3
  [Host]     : .NET 8.0.0 (8.0.23.37506), X64 RyuJIT AVX2
  Job-ELMFFO : .NET 7.0.10 (7.0.1023.36312), X64 RyuJIT AVX2
  Job-VWOUDZ : .NET 8.0.0 (8.0.23.37506), X64 RyuJIT AVX2

InvocationCount=Default  IterationTime=300.0000 ms  MinIterationCount=5  
WarmupCount=6  Quotes=False  Reader=String  

Method	Runtime	Scope	Rows	Mean	Ratio	MB	MB/s	ns/row	Allocated	Alloc Ratio
Sep______	.NET 7.0	Row	50000	5.478 ms	1.00	29	5326.6	109.6	1.14 KB	1.00
Sylvan___	.NET 7.0	Row	50000	6.268 ms	1.14	29	4655.7	125.4	7.18 KB	6.29
Sep______	.NET 8.0	Row	50000	5.184 ms	0.95	29	5629.1	103.7	1.29 KB	1.13
Sylvan___	.NET 8.0	Row	50000	6.061 ms	1.11	29	4814.4	121.2	7.18 KB	6.29
Sep______	.NET 7.0	Cols	50000	7.120 ms	1.00	29	4098.3	142.4	1.14 KB	1.00
Sylvan___	.NET 7.0	Cols	50000	10.867 ms	1.53	29	2685.2	217.3	7.19 KB	6.28
Sep______	.NET 8.0	Cols	50000	6.497 ms	0.91	29	4491.5	129.9	1.3 KB	1.13
Sylvan___	.NET 8.0	Cols	50000	9.941 ms	1.40	29	2935.5	198.8	7.19 KB	6.28
Sep______	.NET 7.0	Asset	50000	55.868 ms	1.00	29	522.3	1117.4	13800.91 KB	1.00
Sylvan___	.NET 7.0	Asset	50000	74.225 ms	1.33	29	393.1	1484.5	14025.46 KB	1.02
Sep______	.NET 8.0	Asset	50000	51.942 ms	0.93	29	561.8	1038.8	13801.23 KB	1.00
Sylvan___	.NET 8.0	Asset	50000	67.061 ms	1.20	29	435.1	1341.2	14025.01 KB	1.02

BenchmarkDotNet=v0.13.5, OS=Windows 10 (10.0.19044.3086/21H2/November2021Update)
AMD Ryzen 9 5950X, 1 CPU, 32 logical and 16 physical cores
.NET SDK=8.0.100-preview.7.23376.3
  [Host]     : .NET 8.0.0 (8.0.23.37506), X64 RyuJIT AVX2
  Job-IEIWJW : .NET 7.0.10 (7.0.1023.36312), X64 RyuJIT AVX2
  Job-VWGRFR : .NET 8.0.0 (8.0.23.37506), X64 RyuJIT AVX2

InvocationCount=Default  IterationTime=300.0000 ms  MinIterationCount=5  
WarmupCount=6  Quotes=False  Reader=String  

Method	Runtime	Scope	Rows	Mean	Ratio	MB	MB/s	ns/row	Allocated	Alloc Ratio
Sep______	.NET 7.0	Row	50000	2.634 ms	1.00	29	11076.8	52.7	1.13 KB	1.00
Sylvan___	.NET 7.0	Row	50000	3.353 ms	1.27	29	8703.7	67.1	7.17 KB	6.33
Sep______	.NET 8.0	Row	50000	2.537 ms	0.96	29	11503.2	50.7	1.13 KB	1.00
Sylvan___	.NET 8.0	Row	50000	3.106 ms	1.18	29	9396.4	62.1	7.17 KB	6.33
Sep______	.NET 7.0	Cols	50000	3.345 ms	1.00	29	8724.8	66.9	1.13 KB	1.00
Sylvan___	.NET 7.0	Cols	50000	5.705 ms	1.71	29	5114.7	114.1	7.18 KB	6.33
Sep______	.NET 8.0	Cols	50000	3.282 ms	0.98	29	8890.3	65.6	1.13 KB	1.00
Sylvan___	.NET 8.0	Cols	50000	5.512 ms	1.65	29	5293.7	110.2	7.18 KB	6.33
Sep______	.NET 7.0	Asset	50000	30.932 ms	1.00	29	943.4	618.6	13799.67 KB	1.00
Sylvan___	.NET 7.0	Asset	50000	43.094 ms	1.39	29	677.2	861.9	14025.03 KB	1.02
Sep______	.NET 8.0	Asset	50000	29.571 ms	0.97	29	986.8	591.4	13799.62 KB	1.00
Sylvan___	.NET 8.0	Asset	50000	30.572 ms	0.99	29	954.5	611.4	14025.23 KB	1.02