Wire calibrated CACF parameters via merged paper refs

[agurgone]

Summary

• Allow paper_parameter_ref blocks to merge resolved registry defaults with local sibling parameter overrides
• Wire FRA CreditAugmentedConsumption to desired_consumption.credit_augmented_v1 while keeping neutral default-consumption settings local
• Move Stage 4 shadow portfolio settings out of asset_returns.paper_stochastic_v1 duplication and into FRA wealth parameter siblings

Validation

• /Users/andone/Documents/python_projects/INET-consumption/.venv/bin/python -m pytest tests/test_macromodel/unit/test_configurations/test_country_config_loader.py tests/test_macromodel/unit/test_configurations/test_households_configuration.py
• /Users/andone/Documents/python_projects/INET-consumption/.venv/bin/python -m pytest tests/test_macromodel/unit/test_agents/test_households/func/test_consumption.py tests/test_macromodel/unit/test_country/test_country.py
• /Users/andone/Documents/python_projects/INET-consumption/.venv/bin/ruff check
• seed-15 t_max=5 smoke completed with OUTPUT_DATA_PATH=/private/tmp/inet-cacf-smoke-output

Smoke diagnostics

• target_consumption_income_growth: active, nonzero=28800/34560
• target_consumption_uncertainty: active with negative calibrated sign, nonzero=28800/34560
• target_consumption_partial_adjustment_gap: active, nonzero=28800/34560
• target_consumption_interest_rate_cashflow: remains zero because consumer_debt_rate_delta is still passed as 0.0

Follow-up

This PR does not wire nonzero real_borrowing_rate or consumer_debt_rate_delta inputs. The interest-rate/cashflow channel remains a separate task.