Add Pipeline3 HG failure analyzer and relax via-limit to reduce circuit100 failures

[0hmX]

Motivation

• Provide a quick way to inspect failing nodes for dataset circuit100 (dataset01) and compute per-node failure probability to guide fixes.
• circuit100 was failing in the Pipeline3 HG high-density phase due to node-level solver exhaustion and a hard ViaPossibilitiesSolver2 via cap.
• Allow targeted tuning (search budget and via allowances) to reduce these failures without broad algorithmic changes.

Description

• Add scripts/analyze-pipeline3hg-failure.ts which runs a scenario (default circuit100), reports failed HighDensity nodes, crossing counts, an estimated per-node probability-of-failure via calculateNodeProbabilityOfFailure, and a suggested remediation plan.
• Expose the script via the npm script analyze:pipeline3hg in package.json so it can be run with bun run analyze:pipeline3hg circuit100.
• Forward pipeline effort into HighDensitySolver by adding effort: cms.effort to the Pipeline3 highDensityRouteSolver constructor params so pipeline-level effort tuning affects high-density budgets.
• Make ViaPossibilitiesSolver2 accept an optional hyperparameter MAX_VIA_COUNT (default remains 5) so callers can relax the via cap when needed.
• Update MultiHeadPolyLineIntraNodeSolver3 to seed ViaPossibilitiesSolver2 with an adaptive MAX_VIA_COUNT based on node complexity (Math.max(5, this.minViaCount * 3, Math.ceil(this.uniqueConnections * 2))) to reduce via-limit failures for complex nodes.

Testing

• Ran bun run analyze:pipeline3hg circuit100 to profile the scenario: before changes the run reported 2 failing nodes (cmn_30, cmn_4) and after the adaptive via-limit change the script reported 1 failing node (cmn_4), indicating reduced via-limit related failures. (script executed successfully).
• Ran bun test tests/pipeline-immutability/autorouting-pipeline3-hg-port-point-pathing.test.ts which passed.
• Ran code formatting via biome format --write on the modified files (format succeeded).

---

Codex Task

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
capacity-node-autorouter	Ready	Preview, Comment	Mar 12, 2026 1:46pm

[tscircuitbot]

🏃 Benchmark This PR

Run benchmarks by commenting on this PR:

/benchmark [solver-name|all] [scenario-limit] --concurrency <n> --effort <n>

Examples:

• /benchmark -> AutoroutingPipelineSolver, all scenarios (default concurrency uses the benchmark runner CPU count)
• /benchmark AutoroutingPipelineSolver -> one solver, all scenarios
• /benchmark all 20 -> all solvers, first 20 scenarios
• /benchmark AutoroutingPipelineSolver 20 --concurrency 8 -> one solver, 20 scenarios, 8 workers
• /benchmark AutoroutingPipelineSolver 20 --effort 2 -> one solver, 20 scenarios, 2x effort

Any PR whose title contains [BENCHMARK TEST] will automatically run the benchmark workflow on PR updates.

[0hmX]

/benchmark

[tscircuitbot]

🏃 Autorouting Benchmark Results

📊 PR Results

Benchmark Results (1x effort)

+---------------------------+----------------+--------------------+-----------+----------+----------+
| Solver                    | Completed %    | Relaxed DRC Pass % | Timed Out | P50 Time | P95 Time |
+---------------------------+----------------+--------------------+-----------+----------+----------+
| AutoroutingPipelineSolver | 83.8% (🕒6.7%) | 1.9% (🕒6.7%)      | 7/105     | 6.4s     | 61.4s    |
+---------------------------+----------------+--------------------+-----------+----------+----------+

Scenarios: 105

📊 Main Branch Results

Benchmark Results (1x effort)

+---------------------------+----------------+--------------------+-----------+----------+----------+
| Solver                    | Completed %    | Relaxed DRC Pass % | Timed Out | P50 Time | P95 Time |
+---------------------------+----------------+--------------------+-----------+----------+----------+
| AutoroutingPipelineSolver | 82.9% (🕒3.8%) | 1.9% (🕒3.8%)      | 4/105     | 6.1s     | 62.9s    |
+---------------------------+----------------+--------------------+-----------+----------+----------+

Scenarios: 105

📦 Artifact: https://github.com/tscircuit/tscircuit-autorouter/actions/runs/23008683422

[0hmX]

/benchmark

[tscircuitbot]

🏃 Autorouting Benchmark Results

📊 PR Results

Benchmark Results (1x effort)

+---------------------------+----------------+--------------------+-----------+----------+----------+
| Solver                    | Completed %    | Relaxed DRC Pass % | Timed Out | P50 Time | P95 Time |
+---------------------------+----------------+--------------------+-----------+----------+----------+
| AutoroutingPipelineSolver | 83.8% (🕒6.7%) | 1.9% (🕒6.7%)      | 7/105     | 6.4s     | 60.6s    |
+---------------------------+----------------+--------------------+-----------+----------+----------+

Scenarios: 105

📊 Main Branch Results

Benchmark Results (1x effort)

+---------------------------+----------------+--------------------+-----------+----------+----------+
| Solver                    | Completed %    | Relaxed DRC Pass % | Timed Out | P50 Time | P95 Time |
+---------------------------+----------------+--------------------+-----------+----------+----------+
| AutoroutingPipelineSolver | 83.8% (🕒2.9%) | 1.9% (🕒2.9%)      | 3/105     | 6.9s     | 61.5s    |
+---------------------------+----------------+--------------------+-----------+----------+----------+

Scenarios: 105

📦 Artifact: https://github.com/tscircuit/tscircuit-autorouter/actions/runs/23039953129

[github-actions]

This PR has been automatically marked as stale because it has had no recent activity. It will be closed if no further activity occurs.

[github-actions]

This PR has been automatically marked as stale because it has had no recent activity. It will be closed if no further activity occurs.