WIP: Render glyphs using canvas

[juancampa]

Uses an OffscreenCanvas to render glyphs instead of the ab_glyph rasterizer.

Benefits:

• We can use OS native fonts that look nicer (i.e. system-ui, SF Mono). We can't embed these due to licensing issues.
• Text shold look sharper since the OS rasterizer should use hints and/or better align outlines to the pixel grid
• Smaller wasm files.

Future work:

• Use OffscreenCanvasRenderingContext2D.measureText to measure text. Consider implementing the ab_glyph::Font trait in a struct that uses wasm_bindgen calls behind the scenes.
• Test with system fonts

Limitations:

• Generally speaking, the texture atlas approach of rendering text is limited by the fact that there's only one version of each glyph (one bitmap), but looking at how the browser renders text, each instance of e.g. the glyph "a", can rasterize to a slightly different bitmap, when they land on different non-integer coordinates. Egui does align each glyph to the pixel boundary in text_layout.rs

---

Status of this PR:

It works but glyphs look unaligned, I think it's because we don't use the canvas to measure the glyphs and the metrics determined by ab_glyph don't match the metrics used by the canvas (canvas bounding boxes tend to be smaller, probably because grid alignment?)

[height]

Link Height tasks by mentioning a task ID in the pull request title or commit messages, or description and comments with the keyword link (e.g. "Link T-123").

💡Tip: You can also use "Close T-X" to automatically close a task when the pull request is merged.