Closes #579.

Summary

This PR clarifies the current DifferenceInDifferences.predict() contract without adding prediction semantics.

predict() is present on the sklearn-like estimator surface, but it currently raises NotImplementedError. The updated docs and error message make that explicit: out-of-sample prediction is unsupported pending a broader post-estimation result-object design, and fitted training-data predictions are available through results_.fitted_values after fit().

Changes

• Update DifferenceInDifferences.predict() docstring and NotImplementedError message.
• Add a regression test for the fitted predict() unsupported contract.
• Add a short API docs note and include predict in the DifferenceInDifferences method autosummary.
• Do not implement prediction or alter result-object semantics.

Validation

• PYTHONPATH=. DIFF_DIFF_BACKEND=python pytest tests/test_methodology_did.py::TestResultsObject::test_predict_contract_points_to_fitted_values -q
• PYTHONPATH=. DIFF_DIFF_BACKEND=python pytest tests/test_methodology_did.py::TestResultsObject -q
• PYTHONPATH=. DIFF_DIFF_BACKEND=python pytest tests/test_methodology_did.py -q
• PYTHONPATH=. DIFF_DIFF_BACKEND=python .venv-py39/bin/python -m pytest tests/test_methodology_did.py::TestResultsObject::test_predict_contract_points_to_fitted_values -q
• python -m ruff check diff_diff/estimators.py
• git diff --check

Docs build note: make -C docs html SPHINXOPTS="-W" currently reaches completion but exits non-zero on pre-existing tutorial 19 DID unknown-target warnings. Those warnings are addressed separately in PR #572.

Security / privacy

Confirmed: no secrets or PII in this PR.