perf(progress): throttle print_progress to 1 Hz instead of every pair

[cbueth]

Previously print_progress was called after every single pair, flushing stdout each time. For a 100-node network that's 9,900 flushes. significant stdbuf overhead.

Replace with rate-limited output: print at most once per second, with a final forced print on the last pair. Applies to both sequential and parallel paths.

Mechanism: track last_print timestamp; only call print_progress when at least 1 second has elapsed (or it's the final pair). Uses Manager.Value in the parallel path for cross-process sharing.

[codspeed-hq]

Merging this PR will improve performance by 18.46%

⚡ 1 improved benchmark
✅ 27 untouched benchmarks

Performance Changes

	Benchmark	BASE	HEAD	Efficiency
⚡	test_ordinal_patterns	15.9 ms	13.4 ms	+18.46%

Tip

Curious why this is faster? Comment @codspeedbot explain why this is faster on this PR, or directly use the CodSpeed MCP with your agent.

---

_{Comparing perf/throttle-progress (6e35d25) with main (875adcd)}