[trajectory-grader] Implement recurrence-laminarity

[Copilot]

Ships recurrence-laminarity (Tier 1, rank 7) from the trajectory graders catalog. The original generating run produced the grader but failed to push, so the change is re-applied here.

LAM isolates stagnation — consecutive steps all matching one earlier fixed state — which none of the built-in graders nor the two already-shipped recurrence graders measure: recurrence-determinism captures ordered repeated subsequences (diagonal structure) and state-revisit-probability-rep only counts revisits.

Changes

• shared/graders/recurrence-laminarity.md (new) — self-contained importable fragment with one graders.recurrence-laminarity entry (unit: ratio, direction: lower_is_better, min: 0.0, max: 1.0) and a pure inline script. Builds the state-recurrence matrix R[i][j] = sequence[i] === sequence[j] (line of identity excluded) over the canonical state sequence and returns vertical-line points (length ≥ 2) / all recurrent points. Column-wise scan, ordered-pair counting on both numerator and denominator — same convention as recurrence-determinism.
• State normalization — reused verbatim from recurrence-determinism: states[].id, states[].firstEventIndex, events[].index, with fallbacks for string states and missing event indices.
• Degenerate inputs — message result for fewer than four state visits; vacuous 0 when no recurrent pairs exist.
• shared/graders/README.md — status flipped to Implemented (7 of 25); no other rows touched.

Behavior

The metric cleanly separates stagnation from ordered repetition:

["a","a","a","a","b","c"]  // → 0.83  stuck on `a`
["a","b","c","a","b","c"]  // → 0     ordered repeat: DET's signal, not LAM's
["a","b","c","a","d","e"]  // → 0     isolated revisit
["a","b"]                  // → not applicable (< 4 visits)

Import with:

imports:
  - shared/graders/recurrence-laminarity.md

• Fixes [trajectory-grader] Implement recurrence-laminarity #55651

[Copilot]

Pull request overview

Adds the Tier 1 recurrence-laminarity trajectory grader for detecting stagnation through vertical recurrence structures.

Changes:

• Implements canonical state normalization and RQA LAM calculation.
• Handles short and non-recurrent trajectories.
• Marks the grader as implemented in the catalog.

Show a summary per file

File	Description
.github/workflows/shared/graders/recurrence-laminarity.md	Defines the importable grader and metric script.
.github/workflows/shared/graders/README.md	Updates implementation status.

Review details

💡 Add a code-review agent skill for context-aware, tailored reviews. Learn more in the docs.

• Files reviewed: 2/2 changed files
• Comments generated: 0
• Review effort level: Balanced

[github-actions]

🎉 This pull request is included in a new release.

Release: v0.87.8