[codex] Model Medicare Part B premiums net of MSP coverage

[MaxGhenis]

What changed

• model baseline medicare_part_b_premiums from enrollment and IRMAA/base premium schedules instead of treating Part B as a copied input
• keep the reported CPS value as medicare_part_b_premiums_reported for audit use
• net modeled Part B out-of-pocket premiums by a cycle-free MSP standard-premium offset
• avoid the prior filing-status cycle in IRMAA calculations by using joint/separate status directly

Why

Part B premiums feed baseline SPM medical out-of-pocket expenses and therefore poverty projections. The prior baseline path copied reported/imputed Part B values, which is especially weak on the PUF/clone side. This PR moves the baseline to a model-consistent Part B path and nets out standard premiums covered on the beneficiary's behalf through MSP-like support.

Impact

• makes baseline SPM MOOP more coherent for projections
• preserves only IRMAA above the MSP-covered standard premium as out-of-pocket
• keeps a separate reported variable for Census-style audits

Validation

• uv run pytest -q policyengine_us/tests/policy/baseline/gov/hhs/medicare/test_part_b_msp_offset.py
• direct local rebuild/measurement showed the MSP offset materially lowers modeled Part B totals but only modestly moves enhanced SPM poverty

Closes #7967.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 100.00%. Comparing base (e8a542a) to head (adcc622).
⚠️ Report is 25 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##              main     #7971    +/-   ##
==========================================
  Coverage   100.00%   100.00%            
==========================================
  Files            1        10     +9     
  Lines           17       134   +117     
==========================================
+ Hits            17       134   +117     

Flag	Coverage Δ
unittests	100.00% <100.00%> (ø)

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

☔ View full report in Codecov by Sentry.
📢 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]

Superseded by clean replacement PR #7972 from codex/medicare-partb-msp-clean.