[kbattocchi]

Summary

Fixes #1066.

Fix two errors in CausalForestDML's doubly robust training-data ATE/ATT inference:

• Count non-missing predictions along the sample axis (axis=0), separately for each outcome-treatment pair and after ATT masking. Previously, pooling counts across outputs or treatment contrasts understated standard errors.
• Preserve outcome-treatment ordering when constructing the doubly robust correction. The previous cross-product/reshape combination mixed the axes when both dimensions exceeded one, affecting point estimates as well as standard errors. Use an explicit einsum with sample, outcome, and treatment axes.

Keep standard-error docstrings focused on the public API. The separately identified multi-output model-selection scoring issue remains out of scope.

Regression coverage

• Deterministic aggregation checks cover vector, single-column, and multi-output outcomes; binary and three-arm treatments; missing rows and unequal column missingness; and masked subsets.
• Fitted-estimator checks cover ATE/ATT estimates, standard errors, confidence intervals, z-statistics, and p-values for every treatment group. They assert finite predictions and use ordinary means, standard deviations, and sample counts, with a comment explaining why the data and estimator seeds must be retained.
• Duplicate-outcome invariance checks compare public ATE/ATT point estimates, standard errors, and confidence intervals before and after duplicating an outcome, for two, three, and four treatment arms. They do not recalculate a reference standard-error formula.

The invariance test exposed the correction-order error. The existing known-ground-truth test uses duplicate outcomes, identical true effects for both non-control treatments, and absolute tolerances that allowed this error to escape; recomputing inference from the estimator's internal corrected predictions also cannot detect incorrect corrections upstream.

Validation

• All 45 focused regression cases plus the existing causal-forest ATE inference test passed (46 total).
• Independently restoring the original missing-axis bug in memory makes all three invariance cases fail.
• Independently restoring the original correction ordering in memory makes the three- and four-arm invariance cases fail; the binary-treatment control still passes.
• Ruff passed for both changed Python files; git diff --check passed.

Local validation uses the repository's last-known-good Cython 3.2.4 build; the original Cython 3.3.0 native-code crash was addressed separately.