Clean package-source Ruff static-analysis findings

[shawcharles]

I noticed a small set of package-source Ruff/static-analysis findings that can be cleaned without changing public API or estimator behaviour.

Current prepared scope:

• declare the ResolvedSurveyDesign forward reference in diff_diff/bootstrap_utils.py under TYPE_CHECKING, removing the package F821 failures;
• remove three no-placeholder f-strings in diff_diff/power.py;
• remove an invalid # noqa: WPS433 marker in diff_diff/spillover.py that Ruff warns about because WPS433 is not a configured Ruff rule;
• remove audited unused locals reported as F841 where the value has no downstream use.

Explicitly out of scope:

• no public API re-export cleanup;
• no TROP import-order / E402 changes;
• no broad import sorting;
• no honest_did.py mathematical variable renames for E741;
• no CI gate.

Validation run locally:

• python -m ruff check diff_diff --select F821,F541,F841,W --output-format=concise
• python3.9 -m py_compile over the touched source files
• PYTHONPATH=. DIFF_DIFF_BACKEND=python pytest tests/test_bootstrap_utils.py tests/test_bootstrap_chunking.py -q
• PYTHONPATH=. DIFF_DIFF_BACKEND=python pytest tests/test_power.py tests/test_methodology_power.py tests/test_spillover.py tests/test_honest_did.py tests/test_methodology_honest_did.py tests/test_imputation.py tests/test_methodology_imputation.py tests/test_prep_dgp_reversible.py tests/test_staggered_rc.py tests/test_staggered.py tests/test_chaisemartin_dhaultfoeuille.py tests/test_dcdh_heterogeneity_cell_period_coverage.py -q

Would you accept a narrow PR with that boundary?