Add default values to physics shapes and update formatting

[aaronfranke]

The goal of this PR is to bring in some of the non-functional changes from #234, making #234 easier to review.

This PR adds default values to the schemas and updates the wording to more closely match KHR/MSFT physics. It also updates the spacing in the Markdown tables to match what's in #234, making that diff easier to review.

This PR contains no functional changes. The default values are already present in the Markdown README file.

I also threw in some lossless optimization of PNG files as a bonus.

[bikemurt]

I've sat in on a couple of these discussions. One thing I noticed in the PR is that the term "full height" is used when describing the height of a capsule, but not used when describing the height of a cylinder. I think this makes sense, since there's no ambiguity in the case of a cylinder. Maybe there's a way to make the wording more consistent between the two.

I think representing the capsule height as the complete height of the capsule is a lot more intuitive for developers (as opposed to the height between the centers of the hemispheres).

It all makes sense to me, I don't have any comments otherwise.

[yankscally]

this is good work as usual,

            "description": "Three number array representing the width, height, and depth of the box (x, y, z).",
            "items": {
                "type": "number",
                "exclusiveMinimum": 0.0
            },
            "minItems": 3,
            "maxItems": 3,
            "default": [ 1.0, 1.0, 1.0 ]
            

I really like this, there is a description for humans, and schemas that could be machine read, hopefully for future applications that may automate how we can use extensions. I would love to see more of this kind of anticipation for using schemas automatically.

[yankscally]

Nice work 👍