Revamp the rabit implementation. (#10112)

This PR replaces the original RABIT implementation with a new one, which has already been partially merged into XGBoost. The new one features:
- Federated learning for both CPU and GPU.
- NCCL.
- More data types.
- A unified interface for all the underlying implementations.
- Improved timeout handling for both tracker and workers.
- Exhausted tests with metrics (fixed a couple of bugs along the way).
- A reusable tracker for Python and JVM packages.

demo/dask/cpu_training.py

@@ -16,7 +16,7 @@ def main(client: Client) -> None:
     m = 100000
     n = 100
     rng = da.random.default_rng(1)
-    X = rng.normal(size=(m, n))
+    X = rng.normal(size=(m, n), chunks=(10000, -1))
     y = X.sum(axis=1)
 
     # DaskDMatrix acts like normal DMatrix, works as a proxy for local