test: per-chip-class estimator accuracy CI gate (~25% median)

## Context

Follow-up from #969 / PR #971. The MoE Tier-2/Tier-1 fixtures and the pre-existing leaderboard replay land, but the issue asked for an offline harness replaying `data/benchmark_cache.json` + `data/community/` with a **per-chip-class ~25% median error** CI gate. That gate was not added (today's overall replay band is still deliberately loose `[0.5, 2.0]`).

## Ask

- Add `llmfit-core/tests/estimator_accuracy.rs` (or extend the existing replay) asserting median |error| ≤ ~25% **per chip class**
- Record baseline medians in the PR that lands it; allow per-class exceptions if a class is already above 25% today rather than turning `main` red
- Keep hermetic (synthetic `SystemSpecs` / embedded cache only)

Related: #969, #971

Start with llmfit-core/tests/estimator_accuracy.rs or the existing leaderboard replay, then inspect data/benchmark_cache.json and data/community/. Confirm how chip classes, synthetic SystemSpecs, and the current overall [0.5, 2.0] replay band are represented. Done means a hermetic CI assertion for median absolute error per chip class, with baseline medians and documented exceptions for classes already above 25%.

Written by the indexing model from the issue text.