Fix Missouri Medicaid parent income limit

[saisharan0103]

Missouri's Medicaid parent/caretaker income limit was modeled through the static statewide FPL ratio in parameters/gov/hhs/medicaid/eligibility/categories/parent/income_limit.yaml, so medicaid_parent_income_limit returned 0.23 for Missouri in every year.

DSS Manual section 1810.020.10 uses a frozen monthly dollar schedule by household size based on the July 16, 1996 AFDC standard. Because poverty guidelines change over time, the effective FPL ratio depends on household size and year.

This change adds a Missouri-specific MHF parent income limit path following the Virginia LIFC precedent: it reads the Missouri monthly household-size dollar schedule from a state parameter, annualizes it with MONTHS_IN_YEAR, and divides by fpg(medicaid_household_size, state_group_str, period, parameters) to preserve the existing FPL-ratio contract.

medicaid_parent_income_limit.formula now returns the Missouri variable for StateCode.MO, while other states, including Virginia's existing override, remain unchanged. Missouri's existing national ratio parameter is kept as fallback historical metadata instead of encoding the frozen dollar schedule into the federal ratio table.

Modified files:

• policyengine_us/variables/gov/hhs/medicaid/eligibility/categories/parent/medicaid_parent_income_limit.py
• policyengine_us/tests/policy/baseline/gov/hhs/medicaid/eligibility/categories/parent/medicaid_parent_income_limit.yaml

Issue: #9122

[codecov]

Codecov Report

❌ Patch coverage is 50.00000% with 11 lines in your changes missing coverage. Please review.
✅ Project coverage is 55.55%. Comparing base (838be3f) to head (5243426).
⚠️ Report is 14 commits behind head on main.

Files with missing lines	Patch %	Lines
...ov/states/mo/dss/mhf/mo_mhf_parent_income_limit.py	57.89%	8 Missing ⚠️
.../categories/parent/medicaid_parent_income_limit.py	0.00%	3 Missing ⚠️

Additional details and impacted files

@@             Coverage Diff              @@
##              main    #9200       +/-   ##
============================================
- Coverage   100.00%   55.55%   -44.45%     
============================================
  Files            2        2               
  Lines           31       36        +5     
  Branches         0        1        +1     
============================================
- Hits            31       20       -11     
- Misses           0       16       +16     

Flag	Coverage Δ
unittests	55.55% <50.00%> (-44.45%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[MaxGhenis]

Cross-family dual review (Claude Fable 5 + GPT-5.6-sol, independent). The dollar schedule itself verifies against the DSS MAGI Appendix A tables (141/241/301/353…616 with the +42 increment), the table is internally monotone with max_household_size consistent, the dispatcher preserves every non-MO state (VA override included), and the size-10 test ratios reproduce exactly from 2025/2026 FPG. Blocked on three things:

1. Wrong issue. The branch name and description cite State tax refunds (SALT refunds) not included in Federal Gross Income / Federal AGI, causing state subtraction distortions #9122, which is about SALT refunds in federal gross income and is untouched here. Please update the description (and don't use closing keywords against State tax refunds (SALT refunds) not included in Federal Gross Income / Federal AGI, causing state subtraction distortions #9122). If there's no open issue for the MO parent limit, say so in the body — the fix can stand on its own citations.
2. Five-point MAGI disregard. The repo's Missouri convention elsewhere folds the conditional 5-point disregard into effective category ceilings (see the comments in young_child/income_limit.yaml and the parent table). mo_mhf_parent_income_limit returns the raw schedule ÷ FPG with no + 0.05. If the 1810.020.10 schedule is a base standard, the effective ratios here are understated by 5 points (0.113 vs 0.163 for the size-10 case). Please establish from the manual (1805.030.20.20.05 covers the disregard) whether the schedule is pre- or post-disregard and align with the existing convention, citing the section.
3. Test coverage. Both cases assert only the intermediate FPL ratio at the unusual size 10. Please add a common household size, an is_parent_for_medicaid-level eligibility case, and an above-table (additional_person) case per the Virginia precedent.