Use cosine similarity in SQL query for embeddings

[Giorgi]

The DiskAnn index is created for vector_cosine_ops but the query was using inner product <#>

[prmerger-automator]

@Giorgi : Thanks for your contribution! The author(s) and reviewer(s) have been notified to review your proposed change.

[learn-build-service-prod]

Learn Build status updates of commit 7985e9f:

✅ Validation status: passed

File	Status	Preview URL	Details
articles/postgresql/azure-ai/generative-ai-azure-openai.md	✅Succeeded

For more details, please refer to the build report.

[v-regandowner]

@mulander

Can you review the proposed changes?

IMPORTANT: When the changes are ready for publication, adding a #sign-off comment is the best way to signal that the PR is ready for the review team to merge.

#label:"aq-pr-triaged"
@MicrosoftDocs/public-repo-pr-review-team

[markingmyname]

#assign: @markingmyname

[markingmyname]

#sign-off

[prmerger-automator]

Invalid command: '#sign-off'. Only the assigned author of one or more file in this PR can sign off. @mulander

[Copilot]

Pull request overview

This PR fixes a bug where the SQL query was using the wrong vector similarity operator. The DiskAnn index is created with vector_cosine_ops for cosine similarity, but the query was incorrectly using the inner product operator.

Changes:

• Changed the vector similarity operator from inner product <#> to cosine similarity <=> in the ORDER BY clause of the embedding similarity search query