feat(math): clojure-legacy blob emission + selection parity — Clojure-exact math_main at step 0

[jucor]

Aligns clojure-legacy emission and selection semantics with Clojure's
prep-main blob (conv_man.clj:45-74), per the step-0 battery diagnosis
(journal 2026-07-22 session 3; fingerprints in docs/divergences.json).
Battery effect: vw:uniform8 (8 steps), vw:single-cut and
biodiversity:uniform8 (8 steps) went from DIVERGENCE to full MATCH;
vw:front-loaded6 remains open (base-cluster count edge at small N).

Emission (Conversation.to_dict, legacy-gated, improved unchanged):

• group-clusters: members emitted as BASE-CLUSTER ids (Clojure folded
  form); unfolded pids stay under the snake alias group_clusters.
• votes-base: per-base-cluster A/D/S bucket lists over sort-by-id
  clustered members (agg-bucket-votes-for-tid, conversation.clj:601-608)
  instead of whole-matrix int totals.
• pca: comment-projection + comment-extremity emitted (existing D12 fns).
• Sign parity: pca.center, base-clusters.x/y, group-clusters centers,
  comment-projection negated at emission — Delphi's matrix is the
  NEGATION of Clojure's (AGREE=+1 vs -1); comps are covariance-derived
  and already equal (verified max|clj+py| = 1e-16 on center).
• repness: {gid: [finalize-cmt-stats-shaped entries]} (repness.clj:173),
  internal dict preserved under repness_full for from_dict round-trip.
• mod-in/mod-out/lastModTimestamp: null until moderation applied.
• tids emitted in Clojure column (first-vote arrival) order with pca
  arrays permuted alongside; from_dict inverts the center sign and
  restores the arrival tracker in legacy mode.

Selection/stat parity:

• group-aware-consensus multiplies (A+1)/(S+2) over EVERY group — a
  zero-S group contributes 1/2, not skipped (conversation.clj:639-641).
• repness rest-stats sum over the OTHER GROUPS only (repness.clj:125),
  excluding unclustered voters from the comparison domain.
• Exact-score ties in repness/consensus selection resolve by Clojure's
  stable sort over named-matrix column order: update_votes tracks
  first-vote arrival order, conv_repness accepts tid_order and the
  selectors iterate as-given in legacy mode.

Also: skip the three math-tree-hashing harness tests in delphi-only CI
images (no /math in the container; run 29881478306's 3 failures).

TDD: tests/test_legacy_blob_shape.py (25 tests, RED observed per fix).

commit-id:0add28f3

---

Stack:

• docs(delphi): overnight pre-cutover audit (s6) — CUTOVER_RUNBOOK + journal #2659
• feat(delphi): zid-sharding for the math poller — one shard = one process #2658
• ci: run postgres integration tests #2637
• feat(delphi): poller-equivalence harness (goal condition 2) — PASSES live on vw + pc-meta-02 #2657
• feat(math): moderation + restart replay parity — mod_update, restart-seam restore, Q13-Q18 #2656
• docs(delphi): R1-parity goal docs, journal (2026-07-18 → 07-24 s5), quirks + divergence ledger #2626
• feat(delphi): Q12 carve-out — pin the cold-tick PCA start to ones in both replay drivers #2655
• feat(math): Q11 — replicate vectorz distance cancellation in legacy kmeans #2654
• feat(math): small-N degenerate parity — run the real math on tiny matrices in clojure-legacy mode #2653
• feat(delphi): Q10 carve-out — pin large-conv cutoffs to certify the deterministic full-PCA path #2652
• feat(delphi): replay loader resolves private datasets under real_data/.local #2651
• feat(math): Clojure hash-map iteration order for the in-conv greedy tie-break (legacy) #2650
• feat(math): clojure-legacy blob emission + selection parity — Clojure-exact math_main at step 0 #2649 ⬅
• feat(delphi): acceptance canonicalization — suppress cross-engine ordering artifacts in certify comparer #2648
• feat(delphi): certify.py battery runner + focuser + fingerprint ledger (Spec A, Phase 0) #2647
• feat(delphi): prodclone extractor — feature-filtered dataset export (Spec B) #2646
• feat(math): un-mirror priority_metric (#2571 resolved) + Q2 prev-tick group-votes #2645
• feat(math): participant-ban leak replication in clojure-legacy mode (Q1) #2644
• feat(delphi): Clojure timing probe (Spec C) #2643
• feat(math): degenerate-tick group-clusterings overwrite + <2-participants edge (legacy mode) #2642
• refactor(math): move impl-flag resolver to polismath/utils/env_flags.py #2641
• feat(delphi): Python math poller service (phase 1) — Clojure math-container replacement #2625
• fix(delphi): Clojure-parity math writers + committing write path in PostgresClient #2624
• feat(math): in-conv greedy floor + persistent carry (PR-E) #2623
• feat(math): Clojure-exact base-cluster lineage + warm start (PR-C) #2622
• feat(math): replay harness H-B — Clojure Mode A driver (dev/replay.clj) + cross-language comparer shim #2621
• feat(math): group-k-smoother in clojure-legacy engine mode (PR-D smoother) #2620
• feat(math): PCA warm start in clojure-legacy engine mode (PR-B) #2640
• feat(math): add POLISMATH_ENGINE_MODE flag + prev-tick warm-start scaffolding (PR-A) #2618
• feat(delphi): replay harness H-A — schedule spec+slicer, Python driver, store, step comparer, CLI #2639
• fix(delphi): cold-start generator copies FULL revote history (no DISTINCT ON dedup) #2638
• fix(math): correct comment-extremity sign in pca_project_cmnts (D1b) #2615
• docs(delphi): add math-algorithm history & routing-regression reference #2613

---

⚠️ Part of a stack created by spr. Do not merge manually using the UI - doing so may have unexpected results.

[Copilot]

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.