Add embed to Index configure calls

[austin-denoble]

Problem

embed was never exposed as an argument for calling configure on IndexResource.

Solution

• Add a new simple ConfigureIndexEmbed(TypedDict) class for representing the argument dictionary shape. I went with this because it aligned with the existing CreateIndexForModelEmbedTypedDict, but I'm not sure if this is best practice in the repo at this point. Maybe a class would be better.
• Update factory, sync, and async resources to pass through embed on configure calls.
• Update legacy Pinecone.configure_index method to support embed.
• Add integration tests to serverless resources to validate converting an existing serverless index to an integrated index using configure or configure_index.

Type of Change

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• This change requires a documentation update
• Infrastructure change (CI configs, etc)
• Non-code change (docs, etc)
• None of the above: (explain here)

Test Plan

New integration tests added. You can pull this branch down and use poetry to run a repl and quickly evaluate things yourself:

poetry run repl

>>> from pinecone import Pinecone
>>> pc = Pinecone(api_key="YOUR_API_KEY")

>>> pc.create_index(name="test-int-inf-convert", dimension=1024, metric="cosine", spec={"serverless": {"cloud": "aws", "region": "us-east-1"}})
{
    "name": "test-int-inf-convert",
    "metric": "cosine",
    "host": "test-int-inf-convert-bt8x3su.svc.preprod-aws-0.pinecone.io",
    "spec": {
        "serverless": {
            "cloud": "aws",
            "region": "us-east-1"
        }
    },
    "status": {
        "ready": true,
        "state": "Ready"
    },
    "vector_type": "dense",
    "dimension": 1024,
    "deletion_protection": "disabled",
    "tags": null
}

>>> pc.db.index.configure(name="test-int-inf-convert", embed={"model": "multilingual-e5-large", "field_map":{"text": "chunk_text"}})
>>> pc.db.index.describe(name="test-int-inf-convert")
{
    "name": "test-int-inf-convert",
    "metric": "cosine",
    "host": "test-int-inf-convert-bt8x3su.svc.preprod-aws-0.pinecone.io",
    "spec": {
        "serverless": {
            "cloud": "aws",
            "region": "us-east-1"
        }
    },
    "status": {
        "ready": true,
        "state": "Ready"
    },
    "vector_type": "dense",
    "dimension": 1024,
    "deletion_protection": "disabled",
    "tags": null,
    "embed": {
        "model": "multilingual-e5-large",
        "field_map": {
            "text": "chunk_text"
        },
        "dimension": 1024,
        "metric": "cosine",
        "write_parameters": {
            "input_type": "passage",
            "truncate": "END"
        },
        "read_parameters": {
            "input_type": "query",
            "truncate": "END"
        },
        "vector_type": "dense"
    }
}

### repeat with async resources / pc.configure_index()

---

• To see the specific tasks where the Asana app for GitHub is being used, see below:
  • https://app.asana.com/0/0/1210417294961252

[rohanshah18]

what's the reason for adding ConfigureIndexEmbed and CreateIndexForModelEmbedTypedDict here?

[austin-denoble]

Adding the classes to _db_control_lazy_imports, which we seem to do for all of our custom models and types. It seemed like CreateIndexForModelEmbedTypedDict was also not included here, so I added it.

This is to allow lazy loaded imports to be exported from the top level of the package here:

pinecone-python-client/pinecone/__init__.py

Line 145 in dfd0125

*list(_LAZY_IMPORTS.keys()),

[rohanshah18]

yeah I was wondering why we added types here but here's the reason why:
#507 (comment)

[rohanshah18]

LGTM! Nice work. Also it makes sense to follow CreateIndexForModelEmbed approach.