autoevals npm package doesn't support `Factuality` with Azure OpenAI

[mongodben]

using autoevals@0.0.119.

not sure why, but when i try to use the Factuality eval scorer with azure openai config, it doesn't work.

get error that API key doesn't exist. error message seems to indicate that it's trying to use the standard OpenAI:

401 Incorrect API key provided: 83be8c47********************6caa. You can find your API key at https://platform.openai.com/account/api-keys.

Error: 401 Incorrect API key provided: 83be8c47********************6caa. You can find your API key at https://platform.openai.com/account/api-keys.
    at APIError.generate (/Users/ben.p/projects/chatbot/node_modules/autoevals/node_modules/openai/error.js:48:20)
    at OpenAI.makeStatusError (/Users/ben.p/projects/chatbot/node_modules/autoevals/node_modules/openai/core.js:263:33)
    at OpenAI.makeRequest (/Users/ben.p/projects/chatbot/node_modules/autoevals/node_modules/openai/core.js:306:30)
    at process.processTicksAndRejections (node:internal/process/task_queues:105:5)

here's my relevant code: https://github.com/mongodb/chatbot/pull/628/files#diff-f687edc4d800857499bfc0a098f9178ee614a6fe107d2611ba38dbb47a93a135R159

i wasn't able to fully discern what was going on by looking at the autoevals code but seems to relate to call to the function cachedChatCompletion(), which the Ragas LLM-as-a-judge scorers don't use.

[ibolmo]

Hi @mongodben I'm looking into this right now! Thanks for reporting.

[ibolmo]

Looks like this may have been fixed by #117. Mind updating to v0.0.120 and reporting back if you're still running into this?

[ibolmo]

Confirmed offline that we're good.