feat: add LEANN connector support

[iamonuwa]

Introduction

To contribute code to the Pathway project, start by discussing your proposed changes on Discord or by filing an issue.
Once approved, follow the fork + pull request model against the main branch, ensuring you've signed the contributor license agreement.

Context

This PR adds a LEANN output connector (pw.io.leann.write) to support LEANN vector indices in RAG pipelines. LEANN is a storage-efficient vector database that uses graph-based selective recomputation to achieve 97% storage reduction compared to traditional solutions.

As discussed in issue #173, LEANN is implemented as a Python output connector following the established pattern from issue #167. The connector accumulates documents and builds the LEANN index on pipeline completion, with support for both HNSW and DiskANN backends.

How has this been tested?

• Unit tests (python/pathway/tests/test_io_leann.py): Test observer behavior including additions, deletions, empty text handling, metadata extraction, and index build logic with mocked LEANN dependencies.
• Integration tests (integration_tests/leann/test_leann.py): End-to-end tests verifying index creation, search functionality, metadata columns, DiskANN backend, custom text columns, directory creation, and edge cases like empty tables.

Types of changes

• Bug fix (non-breaking change which fixes an issue)
• New feature or improvement (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)

Checklist:

• My code follows the code style of this project,
• My change requires a change to the documentation,
• I described the modification in the CHANGELOG.md file.

[CLAassistant]

All committers have signed the CLA.