fix(join): better error when joining empty table

[scbrickley]

Fixes #5134

Checklist

Dear Author 👋, the following checks should be completed (or explicitly dismissed) before merging.

• ✏️ Write a PR description, regardless of triviality, to include the value of this PR
• 🔗 Reference related issues
• 🏃 Test cases are included to exercise the new code
• 🧪 If new packages are being introduced to stdlib, link to Working Group discussion notes and ensure it lands under experimental/
• 📖 If language features are changing, ensure docs/Spec.md has been updated

Dear Reviewer(s) 👋, you are responsible (among others) for ensuring the completeness and quality of the above before approval.

[Marwes]

That is a better error message, but shouldn't joining with an empty table just output an empty table? No need to actually error here?

[Marwes]

I guess we can't construct the type for the output if we try to construct an empty table? (Which is annoying and should be fixed, it is just a bit of a mess)

[scbrickley]

Hmmm, that's a fair point. Maybe I can catch this condition earlier in the transformation and output an empty table before we ever check the column types. I'll tinker with it and get back to you.

[scbrickley]

I guess we can't construct the type for the output if we try to construct an empty table? (Which is annoying and should be fixed, it is just a bit of a mess)

This is the main blocker here. I'm going to merge this PR as is and we can discuss better solutions another time.