Add Azure SQL vector search TypeScript quickstart sample

[diberry]

Azure SQL Vector Search — TypeScript Quickstart Sample

Adds a new TypeScript quickstart sample demonstrating Azure SQL Database's native VECTOR type with Azure OpenAI embeddings.

What's included

• TypeScript sample (�ector-search-query-typescript/) using tedious + Azure OpenAI SDK
• Bicep infrastructure (infra/) — Azure SQL, Azure OpenAI, managed identity, RBAC
• azd-compatible project structure with azure.yaml
• Pre-computed embeddings loaded from JSON data file
• Batched parameterized inserts (10 rows/batch) for efficient data loading
• Two search algorithms: exact nearest neighbor and DiskANN approximate search
• Passwordless authentication via DefaultAzureCredential

Key technical decisions

• Uses tedious native transaction methods (not raw SQL BEGIN/COMMIT) for correct sp_executesql scope handling
• Batched inserts with numbered parameters to minimize round-trips (5 instead of 50)
• Table name validation reserves 13 chars for index naming convention (ix_{table}_embedding)
• Requires Node.js >= 20.6.0 for --env-file flag support
• Uses openai package (not retired @azure/openai) for Azure OpenAI access

Testing

Sample has been verified end-to-end against a live Azure SQL Database with vector search enabled.