feat(propose): --table-grid selects Arm E, absent is off [skip-docs]

The flag threads through `run` and `main` and takes no argument. Arm D's gate
is a run LENGTH where 0 means off; Arm E has no numeric parameter, so a boolean
is the honest shape and an integer would only manufacture a sweepable knob that
means nothing. `run` therefore adds no numeric validation, and the help says
why.

Both prose sites that enumerate the arms `okf build` does not expose are
updated: `src/llm_ingestion_okf/cli.py` and `CLAUDE.md`. The second was found
by review, not by grep of the first -- the same claim lives in two files and
only one of them is code.

The generalised attribution test earned itself in this commit. The first draft
of the Arm E help contained "byte-identical to Arm D -- Arm D rather than Arm
B", and argparse's rendering plus the test's ` --` chunk split meant the
attribution fell OUTSIDE the `table-grid` chunk. The test went red with the
truncated chunk printed, which is exactly the failure it exists to catch: a
whole-output grep would have been satisfied and the attribution would have been
unfindable in the option it belongs to. The clause is now parenthesised.

Arm C's marker check in `tests/test_cli_build.py` gains `rule:table-grid` and
is renamed to speak of all three arms, measured on the artifact rather than on
the flag: a flag `okf build` never passes is not evidence about what it emits.

[skip-docs] is the MEASURED precedent, not a convenience. `grep -c` for
"outline-run", "max-segment-chars", "Arm C" and "Arm D" returns 0 in both
README.md and CHANGELOG.md: an arm flag is documented in its constant's `#:`
comment, in `--help`, and in the round's measurement report, and it is off by
default so it makes no promise to a consumer. `--path-prefix`, which is a real
interface change, does have a CHANGELOG entry. The rule this follows is stated
at docs/2026-09-07-k3-arm-d.md: "interface and behaviour changes yes, arm flags
no." Arm E's report is docs/2026-09-07-k3-arm-e.md, later in this round.

Tests first: 3 red, then green (a fourth, the no-argument test, is honest in
its docstring that it is green before the flag exists too, because argparse
rejects an unknown option with the same code; it becomes evidence only once
the flag is real). 1248 -> 1251.
ruff check: exit 0. ruff format --check: exit 0. mypy --strict src/ tools/:
27 files, Success. pytest -q: exit 0.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

tests/test_cli_build.py

@@ -238,10 +238,12 @@ def test_the_bundle_carries_the_adjudication_layer_the_plans_produce(tmp_path: P
 
 
 def test_arm_c_and_arm_d_are_off_unless_asked_for(tmp_path: Path) -> None:
-    """Both arms stay off by default -- measured on the artifact, not the flag.
+    """Every measurement arm stays off by default -- measured on the artifact.
 
-    `rule:size-split` and `rule:outline` are the markers the two arms write
-    into a plan's `derived` list, so their absence is the arms being off.
+    `rule:size-split`, `rule:outline` and `rule:table-grid` are the markers the
+    three arms write into a plan's `derived` list, so their absence is the arms
+    being off. Named on the artifact rather than on the flag, because a flag
+    `okf build` never passes is not evidence about what it produces.
     """
     inbox = inbox_with_subdirectories(tmp_path)
     plans = tmp_path / "plans"
@@ -253,6 +255,7 @@ def test_arm_c_and_arm_d_are_off_unless_asked_for(tmp_path: Path) -> None:
         text = path.read_text(encoding="utf-8")
         assert "rule:size-split" not in text
         assert "rule:outline" not in text
+        assert "rule:table-grid" not in text
 
 
 def test_segments_off_builds_a_flat_bundle_without_a_bundle_id(tmp_path: Path) -> None:

tests/test_propose_segments.py

@@ -1061,7 +1061,7 @@ def test_a_non_integer_outline_run_is_refused_by_argparse(
 #: Arm C is False on purpose -- its help says "not defined in the K3 method
 #: file" instead, and the point of the check is that each arm's attribution
 #: sits in its OWN chunk.
-ARM_ATTRIBUTION = {"outline-run": True, "max-segment-chars": False}
+ARM_ATTRIBUTION = {"outline-run": True, "table-grid": True, "max-segment-chars": False}
 
 
 def test_each_arm_flag_carries_its_attribution_inside_its_own_option_chunk(
@@ -1483,3 +1483,57 @@ def test_a_split_part_of_a_joined_block_names_all_three_rules_in_order(tmp_path:
         okf_propose_segments.RULE_TABLE_GRID,
         okf_propose_segments.RULE_SIZE_SPLIT,
     ]
+
+
+def propose_arm_e(tmp_path: Path, text: str, name: str = "arm-e-cli.md", *flags: str) -> bytes:
+    """Arm E through the CLI. Bytes, not a dict: the artifact is the promise."""
+    source = write(tmp_path, text, name)
+    out = tmp_path / f"{name}.json"
+    code = okf_propose_segments.main([str(source), "--out", str(out), *flags])
+    assert code == 0, f"proposer exited {code}"
+    return out.read_bytes()
+
+
+def test_the_flag_absent_is_byte_identical_and_present_changes_the_bytes(
+    tmp_path: Path,
+) -> None:
+    """Absent is off; present is not a no-op. Both halves, same text."""
+    absent = propose_arm_e(tmp_path, GRID_TABLE, "absent.md")
+    present = propose_arm_e(tmp_path, GRID_TABLE, "present.md", "--table-grid")
+    assert absent != present
+    # And absent is the pre-Arm-E artifact, not merely "some other bytes":
+    # three entries, none of them naming the join.
+    assert json.loads(absent)["entries"].__len__() == 3
+    assert "rule:table-grid" not in absent.decode("utf-8")
+    assert json.loads(present)["entries"].__len__() == 1
+
+
+def test_arm_d_and_arm_e_are_independent(tmp_path: Path) -> None:
+    """Neither arm may quietly imply the other."""
+    d_only = propose_arm_e(tmp_path, GRID_TABLE, "d.md", "--outline-run", "3")
+    both = propose_arm_e(tmp_path, GRID_TABLE, "both.md", "--outline-run", "3", "--table-grid")
+    assert d_only != both
+    e_only = propose_arm_e(tmp_path, GRID_TABLE, "e.md", "--table-grid")
+    e_zero = propose_arm_e(tmp_path, GRID_TABLE, "e0.md", "--outline-run", "0", "--table-grid")
+    assert e_only == e_zero
+
+
+def test_table_grid_takes_no_argument(tmp_path: Path, capsys: pytest.CaptureFixture[str]) -> None:
+    """Boolean, not an integer pretending to be one.
+
+    Arm D's gate is a run LENGTH and 0 means off; Arm E has no numeric
+    parameter, so inventing one would create a sweepable knob that means
+    nothing. `usage:` in stderr is what separates argparse's refusal from the
+    tool's own `ProposerError` exit, which is also 2.
+
+    Honest about its own limits: this test is green BEFORE the flag exists too,
+    because argparse rejects an unknown option with the same code. It becomes
+    evidence only once `--table-grid` is a real flag, and what it then pins is
+    that the flag stayed boolean.
+    """
+    source = write(tmp_path, GRID_TABLE, "noarg.md")
+    out = tmp_path / "noarg.json"
+    with pytest.raises(SystemExit) as exit_info:
+        okf_propose_segments.main([str(source), "--out", str(out), "--table-grid", "3"])
+    assert exit_info.value.code == 2
+    assert "usage:" in capsys.readouterr().err