docs: classifier (Jev) evaluations can no longer be tested before deploy

[NiveditJain]

Two pages under docs/evaluations/ now describe the opposite of how
classifier (Jev) evaluations behave. Classifier evaluations can no longer be
tested before deploying — they behave like judge evaluations in this respect.

What is wrong

docs/evaluations/jev.mdx, under Testing and backfill:

Unlike a judge, a classifier evaluation can be tested before you deploy
it — test it against real sessions the same way you
would a code evaluation, and read the scores before anything goes live.

This is no longer true. Someone following it gets a refusal from the test
panel, having been told in the docs that it would work.

docs/evaluations/test.mdx does not mention judge or classifier
evaluations at all, so the refusal has no explanation anywhere in the docs.

Suggested wording

jev.mdx — replace the paragraph above with:

Like a judge, a classifier evaluation cannot be tested before you deploy
it. Both spend your organization's model budget, and that spend is only
authorized once the evaluation is deployed. Deploy it to a narrow scope
first and read the early scores instead.

The backfill paragraph that follows it is unaffected and should stay.

test.mdx — add after the intro paragraph:

Code evaluations only. Judge and classifier evaluations cannot be tested
here — both spend your organization's model budget, which is only authorized
once the evaluation is deployed. The panel tells you so if you try.

Worth checking while in there

• The Testing and backfill heading on jev.mdx may read oddly once testing
  is gone from it; Backfill alone may be the better heading.
• Anywhere else that contrasts classifier evaluations with judges on this
  point — the distinction no longer exists.
• The judge page already states the rule correctly and can be the model for
  the phrasing.

[DeepanshuPal]

I can take the docs correction. I'll update jev.mdx and test.mdx with the deploy-before-testing rule and budget explanation you gave, preserve the backfill guidance, and check the surrounding evaluation pages for the same outdated contrast. I'll keep this to documentation and run the repo's docs checks before a PR.