Which issue does this PR close?

Closes #4913.

Rationale for this change

Comet native shuffle currently relies on local shuffle files, which prevents it from being used in deployments backed by a remote shuffle service.

This PR adds Apache Uniffle support while preserving the existing local shuffle path. Comet can push encoded native shuffle blocks to Uniffle and read them back for either regular Spark execution or native shuffle scans.

What changes are included in this PR?

• Generalize the native shuffle writer with pluggable partition-writer backends:
  • LocalPartitionWriter writes local data and index files as before.
  • RssPartitionWriter pushes encoded partition data to a remote shuffle service.
• Add a JNI-based RssPartitionPusher that forwards native shuffle data to the JVM.
• Refactor the Spark-side native shuffle writer and block iterator into reusable interfaces shared by local and remote shuffle implementations.
• Add a comet-uniffle module containing:
  • CometUniffleShuffleManager
  • CometUniffleShuffleWriter
  • CometUniffleShuffleReader
• Support both regular Spark shuffle reads and direct Comet native shuffle scans from Uniffle blocks.
• Add Maven profiles and dependency configuration for building and testing with Uniffle.
• Recognize the Uniffle shuffle manager as a Comet-compatible native shuffle manager.
• Add a dedicated CI job that starts a local Uniffle cluster and runs Comet shuffle tests against it.

Notes: The initial integration supports native shuffle mode only. JVM columnar shuffle with Uniffle is not supported by this PR.

How are these changes tested?

A dedicated GitHub Actions job:

• Starts an Apache Uniffle 0.10.0 coordinator and shuffle server backed by Hadoop 2.10.2.
• Builds Comet for Spark 3.5 with the uniffle profile.
• Runs the Comet Spark test suites using CometUniffleShuffleManager.

Most Comet Spark tests pass with Uniffle. The remaining failures are caused by the current native-only limitation and tests that assume deterministic shuffle ordering without an explicit sort.

Test run: https://github.com/apache/datafusion-comet/actions/runs/29318829876/job/87049939110?pr=4884