run: Support labeled revisions.

[matts1]

This allows users to easily perform comparisons between builds. For example:
jj run -r before=trunk() -r after=@ -- "build && cp out/main /tmp/$JJ_LABEL"

Would generate /tmp/before and /tmp/after

Checklist

If applicable:

• I have updated CHANGELOG.md
• I have updated the documentation (README.md, docs/, demos/)
• I have updated the config schema (cli/src/config-schema.json)
• I have added/updated tests to cover my changes
• I fully understand the code that I am submitting (what it does,
  how it works, how it's organized), including any code drafted by an LLM.
• For any prose generated by an LLM, I have proof-read and copy-edited with
  an eye towards deleting anything that is irrelevant, clarifying anything
  that is confusing, and adding details that are relevant. This includes,
  for example, commit descriptions, PR descriptions, and code comments.

[PhilipMetzger]

LGTM. While the mechanism is great we should find better ways to integrate this cc @hooper for a better opinion and the metadata on snapshot proposal

[josephlou5]

Also what happens if multiple labels apply to some revision, such as -r label1=@ -r label2=@? I don't see a test for this and haven't really dug into the code, but feels like should be documented.

[matts1]

Also what happens if multiple labels apply to some revision, such as -r label1=@ -r label2=@? I don't see a test for this and haven't really dug into the code, but feels like should be documented.

Good point. I'm not yet sure I understand exactly what we want this to look like, so I'll update this then

[matts1]

Looks like the formatter is broken globally

[josephlou5]

Looks like the formatter is broken globally

See #10081.