Address PR #347 R6: drop brittle line ref from StackedDiD; sync llms-full.txt schema

Two P3 cleanups from R6.

P3 #1: the StackedDiD ``target_parameter.definition`` embedded an
internal implementation line reference (``stacked_did.py`` around
line 541). That pointer is not methodology source material and
will go stale under routine estimator edits even when the estimand
itself is unchanged. Removed the reference; definition now stands
on paper/registry terms alone.

P3 #2: ``diff_diff/guides/llms-full.txt`` listed the pre-PR BR/DR
schema top-level keys and omitted ``target_parameter``, so agent-
facing documentation disagreed with the runtime schema. Added
``target_parameter`` to both schema-key lists (BR around line 1779
and DR around line 1844). Documented the field shape
(``name`` / ``definition`` / ``aggregation`` /
``headline_attribute`` / ``reference``), the dispatch tag set, and
the ``headline_attribute=None`` / ``aggregation="no_scalar_headline"``
edge case for the dCDH ``trends_linear=True, L_max>=2`` fit. Also
noted the ``headline.status="no_scalar_by_design"`` value so
guide-driven agents can dispatch correctly. UTF-8 fingerprint
preserved per ``feedback_llms_guide_utf8_fingerprint.md``
(``tests/test_guides.py`` passes).

354 BR/DR + guide tests pass (337 BR/DR + 17 guide). Black clean.

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

Author: igerber

diff_diff/_reporting_helpers.py

@@ -207,15 +207,15 @@ def describe_target_parameter(results: Any) -> Dict[str, Any]:
             "definition": (
                 "The average of post-treatment event-study coefficients "
                 "``delta_h`` (h >= -anticipation), estimated from the stacked "
-                "sub-experiment panel with delta-method SE "
-                "(``stacked_did.py`` around line 541). Each sub-experiment "
-                "aligns a treated cohort with its clean-control set over the "
-                "event window ``[-kappa_pre, +kappa_post]``; each per-horizon "
-                "``delta_h`` is the paper's ``theta_kappa^e`` "
-                "treated-share-weighted cross-event aggregate. The "
-                "``overall_att`` headline is the equally-weighted average of "
-                "these per-horizon coefficients, not a separate cross-event "
-                "weighted aggregate at the ATT level. " + control_clause
+                "sub-experiment panel with delta-method SE. Each sub-"
+                "experiment aligns a treated cohort with its clean-control "
+                "set over the event window ``[-kappa_pre, +kappa_post]``; "
+                "each per-horizon ``delta_h`` is the paper's "
+                "``theta_kappa^e`` treated-share-weighted cross-event "
+                "aggregate. The ``overall_att`` headline is the equally-"
+                "weighted average of these per-horizon coefficients, not a "
+                "separate cross-event weighted aggregate at the ATT level. "
+                + control_clause
             ),
             "aggregation": "stacked",
             "headline_attribute": "overall_att",

diff_diff/guides/llms-full.txt

@@ -1776,11 +1776,23 @@ Schema top-level keys (all always present; missing content uses a
 `{"status": "skipped", "reason": "..."}` shape rather than being absent):
 
 - `schema_version`, `estimator`, `context`
-- `headline`, `assumption`, `pre_trends`, `sensitivity`
+- `headline`, `target_parameter`, `assumption`, `pre_trends`, `sensitivity`
 - `sample`, `heterogeneity`, `robustness`, `diagnostics`
 - `next_steps`, `caveats`, `references`
 
+`target_parameter` (experimental) names what the headline scalar
+represents for each estimator — overall ATT, DID_M, DID_1, cost-
+benefit delta, dose-response aggregate, ASF-based ETWFE, etc.
+Fields: `name` (short stakeholder label), `definition` (full prose),
+`aggregation` (machine-readable dispatch tag, e.g., `"simple"`,
+`"event_study"`, `"delta"`, `"no_scalar_headline"`),
+`headline_attribute` (which raw attribute holds the scalar, or
+`None` when no scalar exists by design — e.g., dCDH with
+`trends_linear=True` and `L_max>=2`), `reference` (citation).
+
 Status enum values: `ran | skipped | error | not_applicable | not_run | computed`.
+`headline.status` also supports `"no_scalar_by_design"` on the dCDH
+no-scalar branch.
 
 ## DiagnosticReport
 
@@ -1830,9 +1842,14 @@ dr.skipped_checks        # dict of {check: plain-English reason}
 ```
 
 Schema top-level keys: `schema_version, estimator, headline_metric,
-parallel_trends, pretrends_power, sensitivity, placebo, bacon,
-design_effect, heterogeneity, epv, estimator_native_diagnostics,
-skipped, warnings, overall_interpretation, next_steps`.
+target_parameter, parallel_trends, pretrends_power, sensitivity,
+placebo, bacon, design_effect, heterogeneity, epv,
+estimator_native_diagnostics, skipped, warnings,
+overall_interpretation, next_steps`. The `target_parameter` block
+mirrors BR's (same shape and dispatch); see BR's section above for
+field semantics including the `headline_attribute=None` /
+`aggregation="no_scalar_headline"` case for dCDH
+`trends_linear=True, L_max>=2` fits.
 
 ### Verdicts and tiers