RAG Example for using REST API in 2.0

[Timoeller]

Is your feature request related to a problem? Please describe.
We want an example of a RAG pipeline + file upload through our REST API.
Making the the app work async (async pipeline + fastAPI) would be the cherry on top of our juicy haystack 2.0 cake.

Since we need updated docker images for starting the REST API, this issue is related to #5990

Describe the solution you'd like
Ideally, we use the existing REST API and just create documentation around it.
See how the indexing pipeline is used we might need a few parameter changes.
Also, the pipeline usage in our search endpoint seems reasonably generic.

[anakin87]

To start, we can simplify the scope of this activity.

I will try to put myself in the shoes of a user who has tried Haystack and wants to start building a real application with it.
I already have some sort of indexing and RAG pipeline.
I want to create an API with two endpoints: uploading/indexing and querying.

The output of this initial activity can consist of:

• a simple page documentation page (with copyable code)
• a demo project/repo with an API example

[r4881t]

Would be useful. I just made this app and its services. If it's any help, here's how my stack looks like

1. Using fastapi to create web server
2. Using sqlalchemy to log the user's query, answer, debug info and other stuff

I may not be able to create a complete example, but if there's a wip, I can contribute to it.

[anakin87]

• The code in https://github.com/deepset-ai/haystack-rest-api is up to date
• The e2e test runs weekly and we get notified if it fails
• My documentation draft is up to date

TO DO:

• Add README to https://github.com/deepset-ai/haystack-rest-api: see Add README haystack-rest-api#5
• Finalize docs

[masci]

Closing as we agreed it's not going to be docs, to be decided whether it's an article or a tutorial. Comprehensive docs about how to deploy pipelines will follow in 2.0