fix: weight:0 un-gates criteria + configurable retry policy

[uipreliga]

Summary

Three independent fixes surfaced by a prior review, plus the code-review follow-ups on top of them.

1. weight: 0 now makes a criterion informational

Previously a weight: 0 criterion contributed nothing to the weighted score but could still single-handedly flip a task to FAILURE — the docs told authors to also set pass_threshold: 0, which was easy to miss.

• New BaseSuccessCriterion.is_gating (weight > 0.0) is the single source of truth for "does a low score here fail the task".
• EvaluationResult.all_criteria_passed and the coder-eval evaluate exit code now gate over is_gating criteria only. A weight: 0 miss no longer flips final_status or the exit code.
• A validator rejects weight: 0 combined with stop_when or suite_thresholds — arming a non-gating criterion for a pass/fail gate (early-stop gate, or the across-row suite gate that drives the run exit code) is incoherent and is now caught at load time.

2. Configurable retry policy

• New RunLimits.retry (RetryPolicy: max_retries / initial_delay / backoff_multiplier, all optional) exposed via -D run_limits.retry.max_retries=0 to fail fast while debugging.
• All retry decisions resolve through one seam, errors/retry.py::resolve_retry_config. A category that is non-retryable in the built-in table (auth/billing) is never made retryable by a per-run override — retryability is a property of the error, not the run.
• Persisted error_details.is_retryable / retry_delay_seconds in task.json now reflect the run's actual policy instead of the built-in defaults.

3. Doc drift + evalboard

• Dropped the non-existent rephrase prompt mutation from the catalog.
• Corrected the stale "max_memory_mb NOT enforced" claim (it maps to docker run --memory; max_cpus/max_pids → --cpus/--pids-limit).
• reports.py failure-reason sampling uses pass_threshold, not score < 1.0.
• evalboard: /trends and /watchlist now ship in the OSS edition (they derive purely from run artifacts); only /path-to-ga stays internal-gated.

Test plan

• make verify green — 3437 passed, 2 skipped, 90.90% coverage, ruff + pyright + custom CE lint clean.
• New regression tests: weight: 0 un-gating semantics, the stop_when/suite_thresholds rejections, retry-policy resolution + fail-fast, error_details policy fidelity, and a coder-eval evaluate exit-code test (with fixture) proving an informational miss doesn't fail the run.

Follow-ups (not in this PR)

Display-only sites (terminal PASS/FAIL label, reports_html, evalboard result pill, checker log line) still label a below-threshold weight: 0 criterion as failed. final_status and both exit codes are correct; a clean fix needs a coordinated gating marker on CriterionResult/CriterionSummary + the evalboard DTO, so it's deferred to avoid partial header-vs-list contradictions.

🤖 Generated with Claude Code

[github-actions]

Claude finished @uipreliga's task in 1m 59s —— View job

---

Code Review In Progress

Review Tasks

• Read review guidelines from .github/code_review.md
• Read project conventions from CLAUDE.md
• Get full PR diff
• Review each changed file for context
• Perform cross-file consistency checks
• Analyze design decisions
• Check for missing elements
• Provide structured feedback

Starting review...