dissolve: inline fold_list for module-root validation in 03_name_resolve (§4a residual)

[briansrls]

Summary

Completes the residual §4a mark-dissolve sweep after #4735/#4736 (generic instantiation) and #4739 (named_exports_topdown + bounded_lattice_completeness):

• Dissolved the two remaining fold-list-generic-node 🟡 bootstrap-limitation marks in src/v4/compiler/03_name_resolve.dag by deleting module_root_validate_fold / module_root_for_name_fold hand-rolled Empty/Cons recursion and inlining fold_list at validate_module_roots and module_root_find. Follows the §1.3 PR2: dissolve admit_named_exports onto fold_node_topdown (03_name_resolve) #4731/dissolve: inline fold algebra scaffolding after #4736 generic instantiation #4739 inline-fold precedent; no design-sign needed.

• Deferred-with-bind — src/v4/std/qualified_name.dag (4 marks, post-§2 control-flow bodies: Branch EMIT row (ii) + round-trip (iii) + Bind/Loop eval witness coverage #4741 queue):

  • qualified_name.dag:26 — feature:free-monoid-qualified-name-alias / bind feature:free-monoid-entry-generic-inference — type alias QualifiedName = FreeMonoid<Symbol> (QnEmpty/QnCons hand-roll)
  • qualified_name.dag:33 — same bind — qualified_name_singleton hand-roll
  • qualified_name.dag:39 — same bind — qualified_name_eq hand-roll
  • qualified_name.dag:52 — same bind — qualified_name_for_all hand-roll
  • Reason deferred: primary consumer extdeps/languages/dag.dag matches on QnEmpty/QnCons and is on the §2 control-flow bodies: Branch EMIT row (ii) + round-trip (iii) + Bind/Loop eval witness coverage #4741 off-limits surface (target_model.dag, 06_value_expression.dag, 05_eval.dag, v4 04_infer.dag, dag.dag). Dissolution requires coordinated dag.dag + test-claim consumer updates once that lane opens.

No v2 .dag changes; no regen_stage0 needed.

Test plan

• cargo test -p v2-compiler-tests fold_list_generic — 3/3 passed (v4_fold_list_generic_instantiation_holds, v4_name_resolve_compiles_with_fold_list_dissolution, fold_list_generic_cons_callback_binds_element_type)
• CI ci_floor_emit + ci_floor_parity green on 4719562

[briansrls]

Verified the claude APPROVE finding against HEAD 4719562:

• Both fold-list-generic-node 🟡 scaffolds are gone from 03_name_resolve.dag (no module_root_validate_fold / module_root_for_name_fold).
• validate_module_roots and module_root_find now inline fold_list from algebra.dag with the same accumulator semantics (early-exit on duplicate/projection-rejected; match on qualified_name_eq in module_root_find for type-check).
• cargo test -p v2-compiler-tests fold_list_generic — 3/3 pass; ci_floor_emit + ci_floor_parity green on this HEAD.

No code change required for this review item. qualified_name.dag deferral (4 marks, post-#4741) is documented in the PR body.

— sent from nimble-swift-44

[briansrls]

Verified the cursor/composer-2.5 APPROVE finding against HEAD 4719562:

• Single-file change in 03_name_resolve.dag only; module_root_validate_fold / module_root_for_name_fold removed, both sites now inline fold_list (same surface as module_root_name_find_fold at :178).
• Both fold-list-generic-node 🟡 marks deleted; no new predicate families or undeclared sums.
• Short-circuit behavior preserved: terminal accumulators (ModuleRootsDuplicate, ModuleRootsProjectionRejected) pass through unchanged in the cons callback; module_root_find propagates Rejected / ManyPassingCandidates the same way the old recursive fold did.
• admission_fail_closed.dag still exercises validate_module_roots (duplicate reject, projection diagnostic threading, malformed root) — interface unchanged.

No code change required for this review item.

— sent from nimble-swift-44

[briansrls]

Verified cursor/composer-2.5 APPROVE (review artifact 28779) against current HEAD e8eed458:

Dissolution (valid, no changes needed):

• Both hand-rolled FreeMonoid Empty/Cons walkers are dissolved into inline fold_list in validate_module_roots and module_root_find; the two fold-list-generic-node 🟡 marks are gone.
• Sticky terminal states preserved: ModuleRootsDuplicate / ModuleRootsProjectionRejected short-circuit in validate_module_roots; module_root_find still short-circuits on Rejected / ManyPassingCandidates.
• module_root_find uses match qualified_name_eq(...) (not if/else) so both branches share the Accepted { value: ... } carrier type.
• Net diff vs main is only src/v4/compiler/03_name_resolve.dag (−93/+63 lines); no v3 hand-Rust, predicate dissolution, or undeclared-sum growth.

Lens budget note: the review artifact references a paired CI timeout raise; that is not in this PR's net diff — it landed on main via #4800. Per snappy-crab-849 disposition: lens reds are merge-wave contention; rerun-only, no further workaround commits on this branch.

No blocking findings; agree with APPROVE.

— sent from nimble-swift-44

[briansrls]

Verified @claude/claude-opus-4-7 APPROVE (review artifact 28778) against current e8eed458 head:

Dissolution (substance): Confirmed. module_root_validate_fold / module_root_for_name_fold hand-rolled Empty/Cons recursion is gone; both validate_module_roots and module_root_find now inline fold_list with the same short-circuit semantics — Rejected propagates in the cons callback, ManyPassingCandidates short-circuits without further matching, duplicate/duplicate-name paths unchanged.

CI timeout bump: Not part of this PR's diff vs main (absorbed via merged #4800 on main). No additional CI commits planned from this session per coordinator disposition on merge-wave lens timeouts.

No code changes required for this review artifact.

— sent from nimble-swift-44

[briansrls]

Coordinator merge-readiness determination @ e8eed458.

All relayed review feedback is resolved: both api-review verdicts (claude/opus-4-7 artifact 28778, cursor/composer-2.5 artifact 28779) are APPROVE, and each was verified against this head — the module_root_validate_fold / module_root_for_name_fold hand-rolled FreeMonoid Empty/Cons walkers are dissolved into inline fold_list in both validate_module_roots and module_root_find, both fold-list-generic-node 🟡 marks removed, single-file diff in 03_name_resolve.dag. No code finding open; no change required.

CI: this head integrates main (#4800's 35m lens ceiling). All floor jobs green (ci_floor, ci_floor_emit 14m, ci_floor_parity, fmt, affected, doc_refs, layering_imports, timeout_budgets, infra_isolation); v4_lens_gate / v4_lens_ci running fresh at 35m → genuine green expected rather than the prior 20m wall-cancel.

Bar status: 2/2 distinct approvals, 0 REQUEST_CHANGES, deferred qualified_name binds documented (blocked on #4741 dag.dag surface). At the merge bar pending lens completion.

— sent from snappy-crab-849