Introduce a non-generic Image base class

[antonfirsov]

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 matches 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

• Implements the proposal demonstrated in API proposal: Pixel-agnostic Image base class #897.
• Change majority of the extension methods working on Image<TPixel> to work on Image instead:

// The non-generic Image.Load overloads no longer return Image<Rgba32>
using (Image image = Image.Load("foo.jpg"))
{
    image.Mutate(x => x.Resize(100, 100);
    image.SaveAsPng(someStream);
}

• Touches Make ImageProcessor-s public #621 by making most of the non-generic processors public. (Exception: Histogram processors, because I'm not entirely happy with all namings.)
• Fix a few doc issues while doing the refactor.
• While implementing better test coverage for Image.Load variants, discovered and added a few missing overloads

Hope I haven't missed any of the important processors and / or extensions!

To enable extensions using color values as parameters on Image, we need to introduce a pixel-agnostic color type. I have a proposal for this, planning to PR it within the next week.

[codecov]

Codecov Report

Merging #904 into master will decrease coverage by 0.31%.
The diff coverage is 95.24%.

@@            Coverage Diff             @@
##           master     #904      +/-   ##
==========================================
- Coverage   89.55%   89.23%   -0.32%     
==========================================
  Files        1031     1060      +29     
  Lines       46115    46028      -87     
  Branches     3261     3268       +7     
==========================================
- Hits        41297    41074     -223     
- Misses       4108     4205      +97     
- Partials      710      749      +39

Impacted Files	Coverage Δ
...ssing/Processors/Filters/AchromatopsiaProcessor.cs	100% <ø> (ø)	⬆️
...s/Convolution/Convolution2PassProcessor{TPixel}.cs	100% <ø> (ø)
.../Processing/Processors/Transforms/SkewProcessor.cs	84.61% <ø> (ø)	⬆️
...ocessing/Processors/Filters/BlackWhiteProcessor.cs	100% <ø> (ø)	⬆️
src/ImageSharp/Image.WrapMemory.cs	100% <ø> (ø)	⬆️
src/ImageSharp/Processing/VignetteExtensions.cs	83.33% <ø> (ø)	⬆️
src/ImageSharp/Processing/DiffuseExtensions.cs	66.66% <ø> (ø)	⬆️
src/ImageSharp/Processing/GlowExtensions.cs	83.33% <ø> (ø)	⬆️
...ocessing/Processors/Filters/TritanopiaProcessor.cs	100% <ø> (ø)	⬆️
...p/Processing/Processors/Filters/InvertProcessor.cs	75% <ø> (ø)	⬆️
... and 291 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 89f3de4...b563167. Read the comment docs.

[codecov]

Codecov Report

Merging #904 into master will increase coverage by 0.04%.
The diff coverage is 94.6%.

@@            Coverage Diff            @@
##           master    #904      +/-   ##
=========================================
+ Coverage   89.55%   89.6%   +0.04%     
=========================================
  Files        1031    1060      +29     
  Lines       46115   46542     +427     
  Branches     3261    3268       +7     
=========================================
+ Hits        41297   41702     +405     
- Misses       4108    4129      +21     
- Partials      710     711       +1

Impacted Files	Coverage Δ
...ssing/Processors/Filters/AchromatopsiaProcessor.cs	100% <ø> (ø)	⬆️
...s/Convolution/Convolution2PassProcessor{TPixel}.cs	100% <ø> (ø)
.../Processing/Processors/Transforms/SkewProcessor.cs	84.61% <ø> (ø)	⬆️
...ocessing/Processors/Filters/BlackWhiteProcessor.cs	100% <ø> (ø)	⬆️
src/ImageSharp/Image.WrapMemory.cs	100% <ø> (ø)	⬆️
src/ImageSharp/Processing/VignetteExtensions.cs	83.33% <ø> (ø)	⬆️
src/ImageSharp/Processing/DiffuseExtensions.cs	66.66% <ø> (ø)	⬆️
src/ImageSharp/Processing/GlowExtensions.cs	83.33% <ø> (ø)	⬆️
...ocessing/Processors/Filters/TritanopiaProcessor.cs	100% <ø> (ø)	⬆️
...p/Processing/Processors/Filters/InvertProcessor.cs	75% <ø> (ø)	⬆️
... and 247 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 44c4763...390bda8. Read the comment docs.

[JimBobSquarePants]

I'm very happy for the most part with this thus far. I'd like to know more about your concerns with the Histogram naming and would like to help there.

[antonfirsov]

AdaptiveHistEqualizationSWProcessor - too many abbrevations.

On the other hand, AdaptiveHistogramEqualizationSlidingWindowProcessor would be way too long. I don't know what is the good solution.

[JimBobSquarePants]

I think we’ve still got a long way to go before we hit the dizzy heights of https://github.com/dotnet/roslyn/blob/master/src/Compilers/CSharp/Portable/BoundTree/BoundTreeRewriter.cs#L100

So I’m ok with the longer form

[antonfirsov]

Ok, went with the longer names. All processors without color parameters should be public (sealed) now.

[antonfirsov]

I'm merging this to go on with the rest of #907. If I made a mistake, I will be able to fix it in the follow up PR-s.