feat: Support embedding dimensions on DeepsetCloudDocumentStore

[dmigo]

Related Issues

• fixes #issue-number

Proposed Changes:

• Add embedding_dim to dc store
• Remove similarity and return_embedding from query params, it is not used

How did you test it?

Notes for the reviewer

Checklist

• [x] I have read the contributors guidelines and the code of conduct
• [ ] I have updated the related issue with new insights and changes
• [ ] I added tests that demonstrate the correct behavior of the change
• [x] I've used the conventional commit convention for my PR title
• [ ] I documented my code
• [ ] I ran pre-commit hooks and fixed any issue

[tstadel]

Removing return_embedding from get_document is fine for now. In the future we might support that in all document stores: https://github.com/deepset-ai/haystack/issues/3007. Currently it's just confusing as it has no effect.

[anakin87]

Sorry for the intrusion. I encountered the same problem with schema generation in the past. I add some information to be helpful...

• the CI shows an error, requesting to locally update and commit the JSON schema
• you try to generate the schema but the right schema is generated only if:
  • you have a full installation of Haystack (pip install -e '.[all]')
  • every module is working and importable (update_json_schema in generate_json_schema.py somewhere tries to import all the possible nodes and I found out that in my installation audio nodes were not working.)

[dmigo]

@anakin87 Thanks for the hints! pip install -e '.[all]' is what I try to run at the moment, however some packages fail to install due to M1 probably. So, I'm debugging this behaviour right now.

[dmigo]

To generate a valid schma I did:

1. brew install openblas
2. brew upgrade cmake
3. Excluded onnx from the list of dependencies
4. GRPC_PYTHON_BUILD_SYSTEM_ZLIB=true OPENBLAS="$(brew --prefix openblas)" pip install -e '.[all]'
5. python .github/utils/generate_json_schema.py