Implement asciidoc support using the asciidork parser crate

[jamesmunns]

This is a draft PR that adds support for asciidoc formatted spec files. It was written considering #185 as an inspiration.

This was largely implemented by claude, I've included the original plan (asciidoc_plan.md) for review, as well as one major rework that was done post-original plan (unhand-plan.md), which removed a hand-rolled parser that claude originally wrote, in favor of using asciidork's parser instead.

I'm going to take this for a spin and see if it generally works, but an initial smoke test shows it is at least partially working:

[jamesmunns]

(I do plan to remove the plan files + squash them out before merging, but I figured they might be useful for reviewing)

[fasterthanlime]

I'm afraid strictdoc got there first and you probably have a heck of a rebase ahead of you! Sorry!

[fasterthanlime]

you probably have a heck of a rebase ahead of you!

You or my claude. We'll see.

[jamesmunns]

Happy to throw your claude or my claude at it, you might be able to guide more intelligently than I can. Definitely still interested in seeing this land, but I did run into some speed bumps of "difference in semantics between markdown and asciidoc" that caused some weirdness while trying to annotate https://hubris.oxide.computer/reference.

The big weirdness is that asciidoc allows for making an "aggregate" doc out of multiple files (sort of like mdbook's SUMMARY.md, but built into the spec), in a way that the current PR doesn't handle well.

There's ALSO some weirdness that asciidoc's = TITLE, which SEEMS similar to markdown's # TITLE, is actually a "document" title, not an <h1>, and the current PR just kinda... ignores that in the rendered output? Unsure exactly what the right way to fix this is.

I don't think this is a blocker, I think if it rebases cleanly it would still be worth merging as-is, and cleaned up/extended in follow-on. We might just not want to start shouting "full support for asciidoc" until we've smoothed down some more edges :)