JPEG : Optimize ScaledCopyTo and Block8x8F for common scale factors

[JimBobSquarePants]

Prerequisites

• I have written a descriptive pull-request title
• I have verified that there are no overlapping pull-requests open
• I have verified that I am following the existing coding patterns and practice as demonstrated in the repository. These follow strict Stylecop rules 👮.
• I have provided test coverage for my change (where applicable)

Description

Improves the performance of the JPEG block scaling and normalization logic in Block8x8F. The main changes are the addition of specialized, loopless scaling routines for common scale factors and the refactoring of the SIMD normalization methods to clarify their responsibilities and remove unnecessary rounding. Fixes #1782

The rounding removal affected our test comparisons far less than I expected but yielded clear visible improvements in some images. For example the replaced image here had blocking that was visible to my eye.

Benchmarks

Performance is sitting pretty following the changes. The playground benchmark has us sitting firmly in 2nd and 3rd place.

BenchmarkDotNet v0.15.8, Windows 11 (10.0.26200.8117/25H2/2025Update/HudsonValley2)
AMD RYZEN AI MAX+ 395 w/ Radeon 8060S 3.00GHz, 1 CPU, 32 logical and 16 physical cores
.NET SDK 10.0.202
  [Host]   : .NET 10.0.6 (10.0.6, 10.0.626.17701), X64 RyuJIT x86-64-v4
  ShortRun : .NET 10.0.6 (10.0.6, 10.0.626.17701), X64 RyuJIT x86-64-v4

Job=ShortRun  IterationCount=5  LaunchCount=1
WarmupCount=5

Method	Mean	Error	StdDev	Ratio
'MagicScaler Load, Resize, Save'	33.30 ms	1.865 ms	0.289 ms	0.17
'ImageSharp TD Load, Resize, Save'	39.29 ms	2.155 ms	0.560 ms	0.20
'ImageSharp Load, Resize, Save'	59.90 ms	1.506 ms	0.233 ms	0.31
'NetVips Load, Resize, Save'	62.45 ms	3.032 ms	0.787 ms	0.32
'SkiaSharp Load, Resize, Save'	77.42 ms	2.112 ms	0.548 ms	0.40
'ImageFree Load, Resize, Save'	133.77 ms	0.498 ms	0.077 ms	0.70
'ImageFlow Load, Resize, Save'	139.14 ms	4.920 ms	1.278 ms	0.72
'System.Drawing Load, Resize, Save'	192.39 ms	2.858 ms	0.442 ms	1.00
'ImageMagick Load, Resize, Save'	207.08 ms	9.209 ms	1.425 ms	1.08