Remove invalid placebo from MultiPeriodDiD handler

run_all_placebo_tests() requires binary time indicator, which
MultiPeriodDiD does not use (it has multi-period time). Removed
the placebo step from _handle_multi_period; HonestDiD (which IS
supported for MultiPeriodDiDResults) remains as the sensitivity tool.

DiD handler already correctly uses _placebo_step() (valid for 2x2
with binary time) and does not suggest HonestDiD (not supported
for DiDResults).

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Author: igerber

diff_diff/practitioner.py

@@ -281,7 +281,7 @@ def _covariates_step() -> Dict[str, Any]:
 def _handle_did(results: Any):
     steps = [
         _parallel_trends_step(),
-        _placebo_step(),
+        _placebo_step(),  # valid: basic 2x2 DiD with binary time
         _step(
             baker_step=4,
             label="Check if data is actually staggered",
@@ -306,7 +306,8 @@ def _handle_multi_period(results: Any):
     steps = [
         _parallel_trends_step(),
         _honest_did_step(),
-        _placebo_step(),
+        # Note: run_all_placebo_tests() requires binary time indicator,
+        # which MultiPeriodDiD does not use. Omit placebo for this type.
         _robustness_compare_step("CS, SA, or BJS"),
     ]
     warnings = _check_nan_att(results)