ContentFit support for images

[Alch-Emi]

Closes #1179
Closes #1014

Adds support for various image fit modes, documented and explained in the new fit module and explained in #1179.

Additionally, the tour example has been updated to show off the new functionality. To this end, an additional image was added to the tour example which satisfies the need for a background (to clearly denote the bounds of the image) and a size smaller than the screen (so that the special behavior of the ScaleDown mode can be seen).

Breaking changes:

• Images now take up space slightly differently, although defaults are a very close approximation of original behavior
• The public fit field was added to the Image struct
• Images on web now follow their height property properly, if set

Non-breaking Changes:

• Addition of ContentFit to iced_core (and appropriate re-exports)
• Update of the tour example
• Addition of the fit(ContentFit) method to the Image widget
• Adjustments to the native rendering of images to support the new ContentFits

Recommended future work:

• This PR does not provide any logic pertaining to the alignment of images within the space alotted to them, and chooses to always align images into the top-left corner, as this behavior is closest to the original behavior. However, varied alignment behavior (or at least the ability to center the content within its frame) will doubtlessly be desired.

fit2.mp4

[hecrj]

This is great! Thank you 🥳

I just have a couple of small suggestions!

[Alch-Emi]

Alright, that should be everything! Any other thoughts?

[Alch-Emi]

Should I include a patch for the iced_pure image widget in this PR too now that that exists, or should that be a separate PR?

[hecrj]

Looks great! 🥳 Thank you 🙇

I made some changes here and there, since I realized we could add ContentFit support for the Svg widget as well!

Let me know what you think and we can merge this! 🚢

Should I include a patch for the iced_pure image widget in this PR too now that that exists, or should that be a separate PR?

Don't worry! The changes there haven't landed yet. I can deal with the conflicts :)

[Alch-Emi]

Looks great! My two concerns with the large & transparent Ferris image are that, because she doesn't have a background, it's hard to tell where the border is, and that it can't showcase the OnlyScaleDown option, since it can't scale up. The border wouldn't normally be a huge deal, and to be fair, it still isn't, but an important part of understanding the image modes is understanding how they affect the border. At the very least, I think we need to use a smaller Ferris, and preferably one with a non-white background.

[hecrj]

I think Contain, Cover, and Fill are easy to grasp.

I have removed the other two from the tour, since we don't really need to be that exhaustive. Maybe we can come up with a dedicated image example to showcase all of the ContentFit variants properly.

Thank you again! 🙇 Let's merge this!