refactor(examples): Update vertical-fl example to Message API and to use flwr-datasets

[jafermarq]

No description provided.

[panh99]

Love this!

[Copilot]

Pull request overview

This PR refactors the vertical-fl example to use the new Message API and integrate flwr-datasets for data partitioning. The refactoring modernizes the example by replacing the old NumPyClient pattern with the Message API and leveraging Flower Datasets' VerticalSizePartitioner for cleaner vertical data partitioning.

Key Changes:

• Migrated from NumPyClient to Message API with @app.query() and @app.train() decorators
• Integrated VerticalSizePartitioner from flwr-datasets to replace custom vertical partitioning logic
• Updated ServerApp to use Grid API for message-based communication
• Removed local CSV file in favor of HuggingFace dataset loading

Reviewed changes

Copilot reviewed 7 out of 7 changed files in this pull request and generated 6 comments.

Show a summary per file

File	Description
vertical_fl/task.py	Replaced custom data processing with flwr-datasets integration and simplified preprocessing; moved ServerModel definition from strategy.py
vertical_fl/strategy.py	Removed file entirely as custom strategy is no longer needed with Message API
vertical_fl/server_app.py	Complete rewrite to use Grid API and Message-based communication for coordinating vertical FL training
vertical_fl/client_app.py	Migrated from NumPyClient to Message API with separate query and train handlers
pyproject.toml	Updated dependency versions and configuration parameters
data/train.csv	Removed local data file (now using HuggingFace dataset)
README.md	Updated documentation to reflect new architecture and usage patterns

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[panh99]

LGTM! Amazing overhaul.