feat: delete all documents for AzureAISearch DocumentStore

[ChinmayBansal]

Related Issues

• fixes add delete_all_documents() operation to AzureAISearchDocumentStore #2311

Proposed Changes:

Added delete_all_documents() method to AzureAISearchDocumentStore with optional index recreation capability.

Implementation details:

• Follows the reference implementation from OpenSearch DocumentStore PR feat: adding the operation delete_all_documents to the OpenSearchDocumentStore #2321
• Supports two deletion strategies via recreate_index parameter:
  • Default (recreate_index=False): Deletes all documents while preserving index (no downtime)
  • Optional (recreate_index=True): Deletes and recreates index with original schema (faster but causes brief downtime)
• Uses Azure Search Client's native delete_documents() and delete_index()/create_index() APIs
• Proper error handling and logging

Side effects addressed:

• Index recreation causes brief downtime but is faster for large datasets
• Document-level deletion preserves index availability but may hit pagination limits for very large indexes (currently supports up to 100,000 documents in single query)

Files changed:

• document_store.py: Added delete_all_documents() method
• test_document_store.py: Added 2 integration tests

How did you test it?

• ✅ Linter: hatch run fmt - All checks passed (16 files unchanged)
• ✅ Type checks: hatch run test:types - No issues found
• ✅ Unit tests: hatch run test:unit - 24 passed
• ✅ Integration tests: Added 2 integration tests:
  • test_delete_all_documents: Verifies deletion of multiple documents
  • test_delete_all_documents_empty_index: Verifies behavior with empty index

Notes for the reviewer

• Implementation follows OpenSearch PR feat: adding the operation delete_all_documents to the OpenSearchDocumentStore #2321 pattern (reference for parent issue #8508)
• The recreate_index parameter addresses the side effects mentioned in the issue
• Default behavior (recreate_index=False) is safer for production as it avoids downtime
• Used Azure's native SearchIndexClient for index operations and SearchClient for document operations
• Pagination limit of 100,000 documents should be sufficient for most use cases; can be adjusted if needed

Checklist

• I have read the contributors guidelines and the code of conduct
• I have updated the related issue with new insights and changes
• I added unit tests and updated the docstrings
• I've used one of the conventional commit types for my PR title: fix:, feat:, build:, chore:, ci:, docs:, style:, refactor:, perf:, test:.