feat: support Anthropic-compatible endpoints for benchmark LLMs

[Sanderhoff-alt]

Add Anthropic as a first-class provider for answer and judge models, validate provider-specific environment variables per role.

[vercel]

Someone is attempting to deploy a commit to the Vectorize Team on Vercel.

A member of the Team first needs to authorize it.

[Sanderhoff-alt]

Hi @kevin-burns,

Thanks for maintaining this repo! I've submitted PR #12 to add support for Anthropic-compatible endpoints.

The only failing check is the Vercel deployment:

Someone is attempting to deploy a commit to the Vectorize Team on Vercel. A member of the Team first needs to authorize it.

Could you please authorize the preview deployment on Vercel? It would be great if we can get this merged.

Appreciate your help!

[nicoloboschi]

can you remove the changes from readme

[Sanderhoff-alt]

can you remove the changes from readme

Sure, I’ll remove the README changes from this PR and keep this PR focused on Anthropic support.

[kevin-burns]

Hi @Sanderhoff-alt : I'm not a maintainer of this repo, just a past contributor (PR #9 added the Ogham memory provider). I don't have permission to authorize Vercel previews here.

The collaborator who merged my PR was @nicoloboschi -- they (and the rest of the @vectorize-io team) are the right folks to ping for the deployment authorization. Tagging them here so they see this thread. Thanks :)