Chore/bump ethrex rev

[jotabulacios]

No description provided.

[jotabulacios]

/bench

[github-actions]

Benchmark — real block (ethrex_mainnet_25368371.bin) (median of 3)

_{continuations · epoch 2^22 · 10 epochs}

Metric	main	PR	Δ
Peak heap	47814 MB	46859 MB	-955 MB (-2.0%) ⚪
Prove time	141.796s	129.424s	-12.372s (-8.7%) 🟢

❓ -8.7% — beyond what 3 runs resolve. Use /bench-abba for a paired test of the same block (default 12 pairs, ~72 min, resolves ~1%).

_{Prove-time spread 1.6% (129.561s / 127.522s / 129.424s)}

_{Commit: 55b400c · Baseline: cached · Runner: self-hosted bench}

[jotabulacios]

/bench-gpu

[github-actions]

GPU Benchmark (ABBA) — a9ec7ce1ee vs main (14 pairs)

_{RTX 5090 · AMD Ryzen 9 9950X 16-Core Processor (32 threads) · Vast.ai datacenter @ $0.6275925925925925/hr · prover/cuda · ethrex real block, continuations · drift-free A/B/B/A}

=== ABBA paired result  (improvement: - = PR faster) ===
  pairs: 14   mean A (PR): 69.187s   mean B (base): 69.096s

  [parametric] paired-t   mean +0.13%   sd 0.84%   se 0.23%
               95% CI: [-0.35%, +0.62%]   (t df=13 = 2.16)
  [robust]     median +0.26%   Wilcoxon W+=62 W-=43  p(exact)=0.5830  (z=+0.56)

  --- server stability (this run; compare across servers) ---
  run-to-run jitter:    A CV 0.70%   B CV 0.61%        (lower = steadier)
  within-session drift: +0.67% over the run, 1st->2nd half +0.10%
    (jitter -> Tier-1 cached gate floor; drift -> whether the cached baseline can be trusted)

  VERDICT: INCONCLUSIVE - effect not separable from 0 at n=14.
           Point estimate ~+0.26% (median). Need more pairs to resolve.

  raw pairs: /tmp/abba_run/pairs.csv

_{- = PR faster. Trust the verdict when paired-t and Wilcoxon agree.}

[diegokingston]

/bench