docs: add Geometry-Aware recentering example (cross-session double dissociation)

[rahimipour-meysam-NeurIPS]

What this adds

A single reproducible example, examples/advanced_examples/plot_geometry_aware_recentering.py, demonstrating unsupervised test-time tangent-space recentering (TangentSpace(tsupdate=True)) for cross-session EEG motor-imagery decoding.

The example reproduces a within/cross double dissociation: recentering gives a clear advantage under CrossSessionEvaluation (there is a between-session covariance shift to correct), while the same two pipelines (recenter on vs. off) are close to indistinguishable under WithinSessionEvaluation (no such shift exists there). This isolates recentering — rather than the choice of final classifier — as the mechanism behind the cross-session gain.

This follows the same pattern as plot_euclidean_alignment.py (#1109): no new class or module is introduced (the mechanism already exists in pyriemann via TangentSpace(tsupdate=True) and, more generally, pyriemann.transfer.TLCenter), just a documented, runnable demonstration using MOABB's own evaluation classes.

Background

The result reproduced here is from a manuscript currently in preparation:

Rahimipour, M., Yang, L., & Van Hulle, M. Simple Geometric Recentering Rivals Deep Sequence Models for Cross-Session EEG Motor-Imagery Decoding. In preparation, 2026.

The full study benchmarks this recentering pipeline against classical Riemannian baselines and deep sequence models (a bidirectional Mamba mixture-of-experts, an SPDNet-style network) across eight public MOABB datasets, with full statistical validation (Friedman omnibus, FDR/Holm-corrected Wilcoxon, Cohen's d, bootstrap CIs, Critical-Difference analysis). This PR only reproduces the core within/cross mechanism on a single dataset, as a self-contained gallery example — not the full benchmark.

I reached out to Sylvain Chevallier by email about this direction and he mentioned PRs to MOABB are welcome, which is what prompted this contribution.

Checklist

• Example script runs end-to-end locally (BNCI2014_001, both CrossSessionEvaluation and WithinSessionEvaluation)
• docs/source/whats_new.rst updated under "Enhancements"
• No new dependencies (pyriemann and moabb evaluation classes only)
• Happy to adjust dataset choice, formatting, or scope per maintainer feedback — this is my first contribution to MOABB and I'd welcome guidance on anything that doesn't match house style.