ci(bench): bump per-target timeout_minutes — current 25/30 cap fails lazy/fast shards

[polaz]

Problem

Today's main pushes (#148, #151, #138) all failed to publish bench data to https://structured-world.github.io/structured-zstd/dev/bench/ because seven bench shards hit the per-target timeout_minutes cap:

Shard	Target	Cap	Outcome
lazy (11 levels: 5..15)	x86_64-gnu	25m	timeout
lazy	i686-gnu	30m	timeout
lazy	x86_64-musl	25m	timeout
fast (8 levels: -7..-1, 1)	x86_64-gnu	25m	timeout
fast	i686-gnu	30m	timeout
fast	x86_64-musl	25m	timeout
btultra2 (3 levels: 20..22)	i686-gnu	30m	timeout

When >=1 bench shard fails, benchmark-aggregate and benchmark-pages never run -> no points published on the dashboard.

Fix

Bump per-target timeout_minutes in .github/workflows/ci.yml bench-matrix:

• x86_64-gnu: 25 -> 45
• i686-gnu: 30 -> 60 (slower 32-bit target)
• x86_64-musl: 25 -> 45

User explicitly chose this minimal-surface fix over splitting heavy shards. Future option (out of scope): split lazy into lazy-lower/lazy-upper and fast into fast-neg/fast-pos so each shard stays under 20 min.

[coderabbitai]

🔗 Related PRs

#75 - fix(ci): publish benchmark delta reports [merged]
#78 - perf(bench): multi-arch relative Rust-vs-FFI dashboard [merged]
#107 - perf(bench): add wildcopy candidate research bench and dashboard range filter [merged]
#142 - ci(bench): cover full -7..=-1, 1..=22 level range with strategy-tag labels [merged]
#144 - fix(bench): restore criterion sample_size to 10 [merged]

---

📝 Issue Planner

_{Check the box below or use the @coderabbitai plan command to generate an implementation plan and prompts that you can use with your favorite coding assistant.}

• Create Plan

---

🧪 Issue enrichment is currently in open beta.

You can configure auto-planning by selecting labels in the issue_enrichment configuration.

To disable automatic issue enrichment, add the following to your .coderabbit.yaml:

issue_enrichment:
  auto_enrich:
    enabled: false

💬 Have feedback or questions? Drop into our discord!