Refactor transcript decoding behind stable events

[chenmingtang830]

Summary

• decode Claude Code and Codex transcripts through a stable harness-event seam
• select one decoder for the full transcript and carry the expected harness through production paths
• expose bounded diagnostics and surface decoder mismatches in cell, CLI, and arena records
• preserve legacy transcript semantics with parity fixtures

Trajectory spike

Tested @letta-ai/trajectory 0.2.0 against three redacted real transcripts from each supported harness. All six failed normalization, so this PR borrows the decode-to-stable-events architecture without adding the dependency.

Validation

• root Vitest: 57 files, 498 tests passed
• arena timeout-prone files rerun serially: 7 files, 46 tests passed
• npm run typecheck
• npm run build

[github-actions]

Normalized records diff

Overall is a macro-average of participating surface scores. Agents are never averaged together.

Overall by agent

Product	Agent	Base pass@1	Head pass@1	Δ	Base pass³	Head pass³	Surfaces
fixture-db	claude-code	92.9%	92.9%	+0.0 pp	78.6% (11/14)	78.6% (11/14)	2
fixture-db	codex	78.6%	78.6%	+0.0 pp	64.3% (9/14)	64.3% (9/14)	2

Surface cells

Product	Agent	Surface	Base pass@1	Head pass@1	Δ	Base pass³	Head pass³	Latency ms	Cost USD	Tokens	Harness version
fixture-db	claude-code	api	85.7%	85.7%	+0.0 pp	71.4% (5/7)	71.4% (5/7)	1100 → 1100	0.4200 → 0.4200	3350 → 3350	2.1.5 → 2.1.5
fixture-db	claude-code	cli	100.0%	100.0%	+0.0 pp	85.7% (6/7)	85.7% (6/7)	1300 → 1300	0.5100 → 0.5100	3720 → 3720	2.1.5 → 2.1.5
fixture-db	codex	api	85.7%	85.7%	+0.0 pp	71.4% (5/7)	71.4% (5/7)	1200 → 1200	— → —	3600 → 3600	0.121.0 → 0.121.0
fixture-db	codex	cli	71.4%	71.4%	+0.0 pp	57.1% (4/7)	57.1% (4/7)	1500 → 1500	— → —	4100 → 4100	0.121.0 → 0.121.0

Operational metrics are context only and never affect ranking.