Add Dreaming-readiness benchmark stage ledger

[yvette-carlisle]

Summary

• add the XY-951 Dreaming-readiness stage ledger in Markdown and machine-readable JSON
• link the ledger from the benchmarking index
• add a regression test that preserves the ledger schema, stage count, typed judgments, baseline counts, and claim boundaries

Benchmark / validation

• jq empty docs/research/2026-06-16-dreaming-readiness-stage-ledger.json
• git diff --check
• cargo test -p elf-eval --test real_world_job_benchmark dreaming_readiness_stage_ledger_preserves_gate_shape -- --exact --test-threads=1
• cargo make fmt
• cargo make lint-fix
• cargo make fmt-check
• cargo make real-world-memory
• cargo make real-world-memory-evolution
• cargo make real-world-memory-consolidation
• cargo make real-world-memory-knowledge
• cargo make real-world-memory-graph-rag
• cargo make real-world-memory-core-archival
• cargo make real-world-first-generation-oss
• cargo make real-world-job-operator-ux
• cargo make checks (nextest: 264 passed, 86 skipped; cargo test also passed with external-service tests ignored as designed)

Baseline judgment

• improved: none
• regressed: none
• unchanged: current-vs-historical correctness, preference evolution, deletion/TTL/tombstone behavior, final competitor retest baseline
• blocked: scheduled-memory-task readiness
• not tested: reviewable consolidation beyond fixtures, memory-summary/top-of-mind live behavior, proactive brief readiness

This is a gate and ledger only. It intentionally does not claim that temporal reconciliation, preference history, consolidation, proactive briefs, scheduled tasks, or competitor adapters are fixed.