Retire v2.std.algebra: list ops to std.algebra, node folds to v2.std.node

[gunbai-bot]

What

Replacement migration (DESIGN §3, delete-first): v2.std.algebra is deleted; no importer, dag-tree or v2, names it any more.

• FreeMonoid operations (fold_list, length, is_empty, contains, list_map, list_head, skip, bag_eq, zip_map, … ) are std.algebra's concept → appended to dag/std/algebra.dag (no new std.list). Re-exports (Cons, Empty, FreeMonoid, Ordering…) are de-forked onto std.algebra.
• Node folds fold_list_node, algebra_empty_conj → v2.std.node.
• CopiedPortFact → its producer v2.lens.cost.copied_port_derivation (consumers: it + copied_port_citations); BoundedLattice*Completeness types → v2.std.bounded_lattice_completeness.
• Dropped: Equivalence, Homomorphism (zero consumers), copied_port_fact_registry_dissolution_note + its dissolution_migration_census residue row.
• ~750 importers rewritten by a name-classified script (import merged into an existing import std.algebra where present, to avoid last-import-wins); qualified v2.std.algebra.f( calls and prose/data citations repointed. Live-path probes that read src/v2/std/algebra.dag now read dag/std/algebra.dag; frontier_ingestion_probe loses its v2_algebra member.

Seed Rust (5 files, what each matched on)

• std_primitive_projection.rs (generated from std.primitive_projection): projection row module_path: "v2.std.algebra" for length → "std.algebra".
• v1_compiler_emit_rust.rs (from 05_emit_rust.dag): owner_module_path == "v2.std.algebra" in rust_call_target_is_v2_algebra_length → "std.algebra" (the call-target identity the ModeledProjection lowering matches).
• v1_compiler_infer.rs (from 04_infer.dag): two note strings only.
• v1_interpreter.rs, required_floor_runner.rs: doc comments only.

Not verified locally

No local build/run (gunbc not runnable on BuildBuddy; whole-tree build is a CI job). Relying on CI for: whole-tree parse, emitted std_algebra.rs compiling with the added generic fns, file_gate ratchet ceilings in accumulator_copy_roster_gate (path now dag/std/algebra.dag), and the native acquisition control. Zero v2.std.algebra references remain except a recorded .diff testdata blob.

Coordination: royal-raven-317 rebases (all declaration moves are owned here). v2.std.collection follows after this lands.

Controls executed (local gunbc built from this branch, cargo build --release -p v1-compiler --bin gunbc, 5m09s)

1. Whole-tree parse — first run at 72d127b REFUSED: module index refused: 1 unparseable .dag source(s) dag/std/algebra.dag: heads-only parse: unterminated function body (my moved block cut skip one line short; this is the CI red). Fixed in cd23e03.
2. Native acquisition at cd23e03: gunbc compile --source-root dag --source-root src/v2 --entry dag/gunbc/machine_intake/ampere_nvparam_console_observation.dag --target dag (formerly import v2.std.algebra { list_reverse }, a dag→v2 edge) → 0 blocking error(s), 450 advisory, compiled: 1 files emitted; the whole index parsed; grep -c v2.std.algebra of the output = 0.
3. Consumer resolution — same compile resolves list_reverse and Cons from std.algebra (consumer above), and v2.std.node.fold_list_node in the diagnostics shows the node fold resolving in its new home.

Re-run at head cd23e03 (both PASS)

• Whole-tree parse: the module index builds over every .dag under dag + src/v2 (it is what refused before the fix); both compiles below passed it (no "module index refused").
• gunbc compile --source-root dag --source-root src/v2 --entry src/v2/compiler/tokenize.dag --target dag — consumer of fold_list, length, is_empty, list_reverse, contains from std.algebra: 0 blocking error(s), 371 advisory, compiled: 1 files emitted, zero v2.std.algebra in output.
• Earlier ampere_nvparam_console_observation.dag compile (formerly refused dag→v2 importer) was also run on the fixed file: 0 blocking, 450 advisory.

Review follow-up receipts (head 7ae3833 after merging main)

• FreeMonoid added to v2.std.node's std.algebra import (196a7ce); all other imports re-merged so no file has two import std.algebra statements (last-import-wins).
• Native route: binary = generation-one v2.cli.compile_cli emitted+built by gunbc test //gunbc/instruments:v2-native-cli (seed gunbc built at the branch head just before the last main merge), sha256 fc30d3bdb2a7ff2d4b1ec09c94d2d8dfd5fb9d8071105178851d45ba3b5be909, closure 44db2256…; instrument exit 0.
  • emit --entry v2.compiler.tokenize --source-root dag --source-root src/v2: acquisition resolved std.algebra (no v2.std.algebra request); LATER UNRELATED refusal: module std.source_annotation … did not front-end: body_lowering_reason_unrecognised_lowered_shape (keen-wolf-69's lane).
  • emit --entry gunbc.machine_intake.ampere_nvparam_console_observation …: formerly refused on v2.std.algebra; now refuses ONLY on v2.std.collection (requested by extdeps.ampere.nvparam) … lies in a tree the requesting tree may not import — the next target, not algebra. NOT a success.
• Seed compile diagnostics, tokenize.dag, same binary/options, base=origin/main vs head: 372 → 371 advisories. By cause: the only net change is −1 unlisted import use on the old src/v2/std/algebra.dag (gone with the file). Every other advisory is re-homed: 11 inhabitance + 14 effect-summary rows moved src/v2/std/algebra.dag → dag/std/algebra.dag; fold_list_node's rows (1 effect + 1 inhabitance) moved to src/v2/std/node.dag. No new cause appears.

[github-actions]

HealPublished repository=gunb-ai/gunbc branch=session/valiant-fox-450 parent=7ae3833c72cb13e82aaee56adb393028027f4057 published_head=ce4fb25ee5b37a32b7b3442aa09cb940746a2289 tree=38888ddffb45140274d196b8c4c855f80dd66c0f paths=src/v1/stage0/src/std_algebra.rs