All `brand.typography.fonts` should be made available in websites

[gadenbuie]

Bug description

When brand.typography.fonts includes additional fonts that aren't used in brand.typography.{base,headings,monospace,...} these additional fonts should still be loaded and made available in websites.

Steps to reproduce

---
title: Load all brand fonts

brand:
  typography:
    fonts:
      - family: Barrio
        source: google
      - family: Roboto
        source: google
    base: Roboto
    # headings: Barrio
---

## Base font

{{< lipsum 1 >}}

## Barrio font

::: {style="font-family: Barrio"}
{{< lipsum 1 >}}
:::

Expected behavior

The second paragraph should be in Barrio font (very distinctive), without Barrio having to appear in any of the family keys in brand.typography.

Actual behavior

The Barrio font isn't loaded and isn't available in the website.

If brand.typography.headings: Barrio, then the Barrio font is loaded, which should not be necessary.

Your environment

• OS: MacOS Sequoia 15.2

Quarto check output

Quarto 1.7.13
[✓] Checking environment information...
      Quarto cache location: /Users/garrick/Library/Caches/quarto
[✓] Checking versions of quarto binary dependencies...
      Pandoc version 3.4.0: OK
      Dart Sass version 1.70.0: OK
      Deno version 1.46.3: OK
      Typst version 0.11.0: OK
[✓] Checking versions of quarto dependencies......OK
[✓] Checking Quarto installation......OK
      Version: 1.7.13
      Path: /Users/garrick/.local/share/qvm/versions/v1.7.13/bin

[✓] Checking tools....................OK
      TinyTeX: (external install)
      Chromium: (not installed)

[✓] Checking LaTeX....................OK
      Using: TinyTex
      Path: /Users/garrick/Library/TinyTeX/bin/universal-darwin
      Version: 2022

[✓] Checking Chrome Headless....................OK
      Using: Chrome found on system
      Path: /Applications/Google Chrome.app/Contents/MacOS/Google Chrome
      Source: MacOS known location

[✓] Checking basic markdown render....OK

[✓] Checking Python 3 installation....OK
      Version: 3.12.2
      Path: /Users/garrick/work/posit-dev/brand-yml/.venv/bin/python
      Jupyter: 5.7.2
      Kernels: python3

[✓] Checking Jupyter engine render....OK

[✓] Checking R installation...........OK
      Version: 4.4.2
      Path: /Library/Frameworks/R.framework/Versions/4.4-arm64/Resources
      LibPaths:
        - /Users/garrick/Library/R/arm64/4.4/library
        - /Library/Frameworks/R.framework/Versions/4.4-arm64/Resources/library
      knitr: 1.49
      rmarkdown: 2.29

[✓] Checking Knitr engine render......OK

[gadenbuie]

Another way this would fail is to include Barrio in headings.family, but with fallback fonts:

brand:
  typography:
    fonts:
      - family: Barrio
        source: google
      - family: Roboto
        source: google
    base: Roboto
    headings: Barrio, Times New Roman

[gordonwoodhull]

Typst passes the first test, since caching of font resources is independent of what is used.

But it fails the second test due to not anticipating a list. 🤦

Producing

#show heading: set text(font: "Barrio, Times New Roman", )

which should be

#show heading: set text(font: ("Barrio", "Times New Roman"))

## Heading is `#context text.font`{=typst}

produces

But it's really Typst's global fallback font, Linux Libertine. 😝

Very important for multilingual text, will fix and backport for Typst.

(moved to #11933)

[gordonwoodhull]

Currently font imports (whether from google, bunny, or file) happen during resolveHTMLFontInformation:

quarto-cli/src/core/sass/brand.ts

Lines 464 to 468 in eb425ff

	const font = getFontFamilies(family);
	result.family = resolveGoogleFontFamily(font) ??
	resolveBunnyFontFamily(font) ??
	resolveFileFontFamily(brand, font) ??
	family;

resolveHTMLFontInformation is called for each typography element.

For each font family specified for a typography element, we

• filter the fonts resources by the font family name
• try to load those filtered font resources by google, bunny, or file

Each of the resolve functions returns either undefined or the font family name. They don't currently change the font family name, so this is equivalent to returning boolean. There is also some logic in each resolve function to double check that font family names are consistent, but I think this is unnecessary because the fonts resources were already filtered by family name.

Given that resolution falls back to the original specified font family name, and resolution does not change the font family name, I think the code quoted above has no direct effect. It only has side effects.

So I think we could resolve directly the fonts resources outside of typography elements, and remove this logic from here. The net effect would be to generate imports for every font in fonts, without any prioritization of one source over another, and without filtering by the family names that are used by typography elements. (Which is the initial issue raised here.)

However, the existing logic is sufficiently complicated that I'm not sure I want to do this as a backport. Either I am missing some complication about font family names, or this code could be simplified.