Add data ingestion quickstart for processing custom data

[Copilot]

Summary

Adds quickstart documentation for Microsoft.Extensions.DataIngestion library, demonstrating complete ETL pipeline for RAG scenarios.

Contributes to #50534

Changes

Documentation

• New quickstart: docs/ai/quickstarts/process-data.md
  • Document reading with MarkdownReader
  • AI-powered enrichment (image alt-text, summaries)
  • Semantic chunking with embedding-based similarity
  • Vector storage using SQLite
  • Interactive semantic search
• Dual-platform support via zone pivots (OpenAI/Azure OpenAI)
• Added to "Chat with your data (RAG)" section in TOC

Code Snippets

• Complete compilable C# projects for both platforms
• Demonstrates pipeline composition: reader → enricher → chunker → writer
• Includes sample data and region markers for doc references

// Compose data ingestion pipeline
using IngestionPipeline<string> pipeline = new(reader, chunker, writer, loggerFactory)
{
    DocumentProcessors = { imageAlternativeTextEnricher },
    ChunkProcessors = { summaryEnricher }
};

await foreach (var result in pipeline.ProcessAsync(new DirectoryInfo("./data"), searchPattern: "*.md"))
{
    Console.WriteLine($"Completed processing '{result.DocumentId}'. Succeeded: '{result.Succeeded}'.");
}

Based on sample from https://github.com/luisquintanilla/DataIngestion and blog announcement at https://devblogs.microsoft.com/dotnet/introducing-data-ingestion-building-blocks-preview/

Warning

Firewall rules blocked me from connecting to one or more addresses (expand for details)

I tried to connect to the following addresses, but was blocked by firewall rules:

• devblogs.microsoft.com
  • Triggering command: /usr/bin/curl curl -s REDACTED (dns block)
  • Triggering command: /usr/bin/wget wget -q -O /tmp/blog.html REDACTED (dns block)

If you need me to access, download, or install something from one of these locations, you can either:

• Configure Actions setup steps to set up my environment, which run before the firewall is enabled
• Add the appropriate URLs or hosts to the custom allowlist in this repository's Copilot coding agent settings (admins only)

Original prompt

Add a quickstart or tutorial for processing custom data based on the data ingestion sample described in https://devblogs.microsoft.com/dotnet/introducing-data-ingestion-building-blocks-preview/. It should live under the "Chat with your data" section of the AI TOC and be titled "Process data" or something similar. This task contributes to #50534.

---

💬 We'd love your input! Share your thoughts on Copilot coding agent in our 2 minute survey.

---

Internal previews

📄 File	🔗 Preview link
docs/ai/quickstarts/process-data.md	Process custom data for AI applications
docs/ai/quickstarts/snippets/process-data/data/sample.md	docs/ai/quickstarts/snippets/process-data/data/sample
docs/ai/toc.yml	docs/ai/toc

[Copilot]

Pull request overview

This pull request adds a new quickstart tutorial for the Microsoft.Extensions.DataIngestion library, demonstrating how to build an ETL pipeline for RAG scenarios. The quickstart shows users how to read Markdown documents, enrich them with AI, chunk them semantically, and store them in a vector database for semantic search. Additionally, the PR includes cleanup changes to other quickstart files, removing hardcoded model names from user secrets in favor of inline string values.

Key Changes

• New quickstart documentation showing end-to-end data ingestion pipeline for AI applications
• Sample code demonstrating pipeline composition with readers, enrichers, chunkers, and vector storage
• Code cleanup across existing quickstarts (text-to-image, structured-output) to simplify configuration

Reviewed changes

Copilot reviewed 10 out of 10 changed files in this pull request and generated 5 comments.

Show a summary per file

File	Description
docs/ai/toc.yml	Adds new quickstart entry under "Chat with your data (RAG)" section
docs/ai/quickstarts/process-data.md	New quickstart documentation for data ingestion pipeline
docs/ai/quickstarts/snippets/process-data/Program.cs	Complete C# example implementing data ingestion with Azure OpenAI
docs/ai/quickstarts/snippets/process-data/ProcessData.csproj	Project file with required NuGet packages
docs/ai/quickstarts/snippets/process-data/data/sample.md	Sample Markdown document for testing the pipeline
docs/ai/quickstarts/text-to-image.md	Removed unnecessary model name from user secrets configuration
docs/ai/quickstarts/structured-output.md	Removed unnecessary model name from user secrets configuration
docs/ai/quickstarts/snippets/text-to-image/azure-openai/Program.cs	Hardcoded model name instead of reading from user secrets
docs/ai/quickstarts/snippets/structured-output/Program.cs	Hardcoded model name instead of reading from user secrets
docs/ai/how-to/snippets/access-data/ArgumentsExample.cs	Hardcoded model name instead of reading from user secrets

[BillWagner]

I had one comment on the code, and then this is ready to merge.

[luisquintanilla]

Looks good. Added a few comments