fix(ci): bump AVM check-circuit per-tx timeout 30s -> 60s

[AztecBot]

What

Fixes the nightly/next AVM Circuit Inputs Collection and Check workflow (avm-check-circuit job), which has been failing on every run.

Root cause

avm_check_circuit_cmds in yarn-project/end-to-end/bootstrap.sh runs bb-avm avm_check_circuit on every dumped e2e AVM tx with a per-command budget of TIMEOUT=30s, fanned out via parallelize with --halt now,fail=1.

A single tx — e2e_multiple_blobs tx 0x242e67c9... — consistently takes 34–35s and hits the 30s timeout (exit code 124). That one timeout makes parallelize abort the whole job, so the GitHub step exits 124 even though all other ~105 txs pass in 4–11s.

This is not a flake: the same tx timed out in both the latest run (35s, run 27860137759) and the prior run (34s, run 27805430170). e2e_multiple_blobs is deliberately heavy — it deploys the large AvmTest/AvmGadgetsTest contracts and emits more than a blob's worth of data, so its AVM circuit is bigger than the others and sits just over the 30s line (worsened by CPU contention during the parallel run).

Fix

Raise the per-command budget from 30s to 60s — roughly 2× the observed ~35s, with headroom for parallel-load variance — and refresh the now-stale comment that claimed "all of the e2e test txs seem to be relatively small". No behavior change for the passing txs; the slow tx now completes within budget.

---

Created by claudebox · group: slackbot