fix: mark epoch-specific validators doctest as no_run

[r-near]

Summary

• Mark the EpochReference::EpochId doctest in src/methods/validators.rs as no_run to prevent flaky CI failures

Problem

The first doctest queries validators by a hardcoded epoch ID against a live RPC endpoint. NEAR nodes (including archival nodes) only retain ~3 epochs (~36 hours) of validator/epoch data in the EpochManager, so any hardcoded epoch ID inevitably goes stale, causing cargo test to fail with UnknownEpoch.

Fix

Changed the code fence from ``` to ```no_run so the example still compiles (ensuring correctness) but doesn't execute against a live endpoint during tests. This matches the established pattern used in other doctests throughout this repo (broadcast_tx_async.rs, broadcast_tx_commit.rs, tx.rs, send_tx.rs, experimental/protocol_config.rs).

The second doctest (EpochReference::Latest) remains runnable since it always queries the current epoch.

[Copilot]

Pull request overview

Marks a network-dependent doctest in the validators method docs as no_run so CI doesn’t flake when a hardcoded EpochId becomes unavailable on archival RPC nodes.

Changes:

• Updated the EpochReference::EpochId doctest fence in src/methods/validators.rs to ```no_run so it compiles but does not execute.

---

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