docs: remove unreleased research material
docs: explain composition mechanisms Publish the fixed-membership evidence for BFS locality and CC work reduction, together with the unresolved PR and CC-SV boundaries. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
docs: align wiki with frozen evidence Publish the GORDER-primary performance story, sealed composability evidence, Compact-and-Emit documentation, and explicit selector/Rabbit limitations while removing retired pages. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
docs: refresh versioned figure embeds Append the graphbrew-public-v3 cache key to every raw SVG URL so the GitHub wiki displays the current correlated figures instead of a stale CDN response. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
docs: publish correlated GraphBrew figure narrative Add the six-stage running example, measured catalog strips, current-versus-proposed claim boundary, reproducible figure links, and synchronized mechanism documentation. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Add editable reordering figure catalog Publish one transformation figure per algorithm ID and link Lucidchart-importable draw.io sources and the multi-page editable bundle. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Explain the low-reuse policy and cost controls Add the new-graph decision, Rabbit fallback interpretation, evidence-accounting boundary, and detailed explanations for cd_parallel, sgmb4096, gordf5000, and norefine. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Clarify GraphBrew composition and runtime policy Replace the generated-chart narrative with one scientific architecture, distinguish explicit compositions from the frozen deterministic selector, and retain ML surfaces only as labeled history. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Document all-kernel low-reuse selector Publish the validated reuse-1/2 feature rule, exact cost-matched composition, Boost fallback, and held-out evidence scope. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Document Gorder8 cost attribution Clarify that the non-Rabbit quality winner requires full Leiden and that the measured one-pass simplification loses kernel quality. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Bind ordering guidance to measured evidence Replace universal when-to-use recommendations with objective-scoped claims, document non-Rabbit COMPOSE wins, and add the reproducible experiment contract. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Publish GraphBrew wiki