Address PR #356 CI review round 12 (1 P2 guide)

Balanced-panel eligibility gate tightened. `PanelProfile.is_balanced`
is computed from the unique `(unit, time)` support, so it stays `True`
even when duplicate rows exist — `duplicate_unit_time_rows` is a
separate alert for that case. But ContinuousDiD / EfficientDiD /
HeterogeneousAdoptionDiD all require exactly one observation per cell:
EfficientDiD and HAD raise ValueError on duplicates at fit() time, and
ContinuousDiD's precompute path silently resolves duplicates via
last-row-wins (which can change the estimand without warning).

Guide §3 balanced-panel-eligibility block now requires BOTH
`is_balanced == True` AND absence of the `duplicate_unit_time_rows`
alert before routing to these estimators, with the specific failure
mode (raise vs silent overwrite) named per-estimator and a concrete
two-step fix (`balance_panel()` + `drop_duplicates([unit, time])`).

Tests: extended the semantic guide test to assert the guide mentions
`duplicate_unit_time_rows` and uses "BOTH/both" wording around the
is_balanced gate, so future edits cannot silently drop the duplicate-
row half of the eligibility requirement.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

Author: igerber

diff_diff/guides/llms-autonomous.txt

@@ -269,15 +269,24 @@ supported / out of scope; `warn` supported but with documented caveats;
   raises `ValueError`. For full staggered support that retains
   every cohort, use `ChaisemartinDHaultfoeuille` instead.
 
-**Balanced-panel eligibility.** The following estimators hard-reject
-unbalanced panels (each raises `ValueError` at `fit()` when a unit is
-missing any period): `ContinuousDiD`, `EfficientDiD`, `SyntheticDiD`,
-`HeterogeneousAdoptionDiD`. Gate these on
-`PanelProfile.is_balanced == True`; if `False`, pre-process with
-`diff_diff.prep.balance_panel()` or pick a balance-tolerant
+**Balanced-panel eligibility.** The following estimators require
+exactly one observation per `(unit, time)` cell with every unit
+observed in every period: `ContinuousDiD`, `EfficientDiD`,
+`SyntheticDiD`, `HeterogeneousAdoptionDiD`. Gate these on BOTH
+`PanelProfile.is_balanced == True` AND the absence of the
+`duplicate_unit_time_rows` alert (`is_balanced` is computed from the
+unique-key support and stays `True` when duplicates exist; the
+alert is the separate signal for duplicates). Treat both
+conditions as hard gates: `EfficientDiD` and
+`HeterogeneousAdoptionDiD` raise `ValueError` at `fit()` on
+duplicate cells, and `ContinuousDiD`'s precompute path resolves
+duplicates with last-row-wins (silent overwrite that can change
+the estimand). If either condition fails, pre-process with
+`diff_diff.prep.balance_panel()` and a
+`drop_duplicates([unit, time])` pass, or pick a balance-tolerant
 estimator from the remaining rows (CS/SA/dCDH/Imputation/TwoStage/
-Stacked/ETWFE all accept unbalanced input, with some caveats in their
-own docs).
+Stacked/ETWFE all accept unbalanced input, with some caveats in
+their own docs).
 
 
 ## §4. Estimator-choice reasoning by design feature

tests/test_profile_panel.py

@@ -675,6 +675,22 @@ def test_guide_api_strings_resolve_against_public_api():
         "for full staggered support"
     )
 
+    # Balanced-panel gate is incomplete with `is_balanced` alone because
+    # duplicate (unit, time) rows don't flip is_balanced. Guide must
+    # require BOTH is_balanced == True AND absence of the
+    # duplicate_unit_time_rows alert before routing to the duplicate-
+    # intolerant estimators (ContinuousDiD silently overwrites
+    # duplicates via last-row-wins; EfficientDiD/HAD raise).
+    assert "duplicate_unit_time_rows" in text, (
+        "Guide must name the duplicate_unit_time_rows alert as part of "
+        "the balanced-panel eligibility gate"
+    )
+    assert "BOTH" in text or "both" in text, (
+        "Guide must require BOTH is_balanced and absence of the "
+        "duplicate_unit_time_rows alert before routing to duplicate-"
+        "intolerant estimators"
+    )
+
 
 def test_min_pre_post_use_per_unit_observed_support():
     """On an unbalanced panel where one treated unit is missing its