Adding support for non-Azure openai instances

[carlotta94c]

Purpose

This PR adds OpenAI API support for users who don't have access to Azure OpenAI. Additions:

• Calling OpenAI endpoint if env variable OPENAI_API_TYPE is set to openai (default models used: text-davinci-003, gpt-3.5-turbo and text-embedding-ada-002)
• Making bicep configuration conditional (Azure OpenAI resource is created only if OPENAI_API_TYPE env variable is set to 'azure')
• Adding instruction to switch from an Azure OpenAI endpoint (default) to a non-azure OpenAI instance

Does this introduce a breaking change?

[ ] Yes
[x ] No

Pull Request Type

What kind of change does this Pull Request introduce?

[ ] Bugfix
[ x] Feature
[ ] Code style update (formatting, local variables)
[ ] Refactoring (no functional changes, no api changes)
[x ] Documentation content changes
[ ] Other... Please describe:

How to Test

• Get the code

git clone [repo-address]
cd [repo-name]
git checkout [branch-name]
npm install

• Test the code
  Configure an existing non-Azure OpenAI instance and deploy the app:

1. Run azd env set OPENAI_API_TYPE openai
2. Run azd env set OPENAI_ORGANIZATION {Your OpenAI organization}
3. Run azd env set OPENAI_API_KEY {Your OpenAI API key}
4. Run azd up
5. Test the app by running it locally (Run Task -> Start App in VSCode) and in the Cloud (click on the link printed in the console after deployment)

What to Check

Verify that the following are valid

• Running 'azd up' without changing env variables settings described above will ensure the same behavior of provisioning and deployment as before this PR (using Azure OpenAI resource is the default).

Other Information

[bgadrian]

Is there an extra config step is being missed out from the readme? openai API responds with an error "that the model is not specified", when the embeddings are done.

[carlotta94c]

Is there an extra config step is being missed out from the readme? openai API responds with an error "that the model is not specified", when the embeddings are done.

@bgadrian thanks for the heads up! Yes I did miss the configuration of the openai embedding model name. I added it into the bicep file.

[pamelafox]

@carlotta94c I've merged main, made some style changes, verified that Azure OpenAI still works, and added unit tests (which resulted in snapshot updates due to the paramaterization). Please verify that non-Azure OpenAi deployment still works for you.

[carlotta94c]

@carlotta94c I've merged main, made some style changes, verified that Azure OpenAI still works, and added unit tests (which resulted in snapshot updates due to the paramaterization). Please verify that non-Azure OpenAi deployment still works for you.

@pamelafox I've add some fixes and tested again deploy and run with non-Azure openai endpoint. It's working well now!

[pamelafox]

Okay, there's still a non-optimal user experience for non-Azure OpenAI users where they get prompted for the location, but everything seems to work. Thanks for all the work on this Carlotta.