Keep safe-layer DRC improvements

[ShiboSoftwareDev]

Issue

The portfolio rejects a safe trace-layer candidate whenever any trace or pad error remains, even when exact DRC shows that the candidate is a strict improvement.

Implementation

Accept the candidate when the existing exact DRC comparison says it is better and its via-related DRC count did not increase.

The same snapshot now shows 4 DRC errors at the input, 1 in the candidate, and 1 in the portfolio output.

Checks

• bun test (55 passed)
• bun run formatcheck
• bun run typecheck

[tscircuitbot]

Benchmark This PR

Run benchmarks by commenting on this PR:

/benchmark [benchmark.sh args...]
/benchmark [scenario-limit|all] --dataset <drc14|srj18> --concurrency <n> --effort <n> --max-iterations <n>

Everything after /benchmark is forwarded directly to ./benchmark.sh.

Examples:

• /benchmark -> run default SRJ18 benchmark settings
• /benchmark 10 -> run first 10 SRJ18 samples
• /benchmark --dataset srj18 -> run all SRJ18 samples
• /benchmark 1 --dataset srj18 -> run the first SRJ18 sample
• /benchmark --dataset drc14 -> run all DRC14 samples
• /benchmark all --concurrency 4 -> run all SRJ18 samples with 4 workers
• /benchmark --scenario-limit all --effort 2 -> run all samples with effort 2
• /benchmark --scenario-limit 20 --max-iterations 100 -> run 20 samples with a max iteration override

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