Build the complete application

## User
A national-accounts researcher studying measurement uncertainty.

## Build
Track the gap between expenditure-side and income-side output estimates. Explain revisions and source differences without forcing the two measurements to agree.

## Input Contract
- Key: quarter + release_id + measure.
- Fields: level, unit, annual_rate, methodology_version, source_notes.
- Define the displayed gap explicitly as GDP minus GDI. Do not mix releases without labeling.

## Workflow
1. Align measurement periods and release versions.
2. Calculate the signed gap and its revision history.
3. Review source-note changes against the measurement bridge.
4. Export a research note that preserves unresolved differences.

## Interface
Paired output series, a signed-gap chart, and a revision explanation workspace.

## Deliverables
- A GDP-GDI vintage comparison store.
- An evidence-bound measurement-gap report.
- Runnable local application, setup commands, and a dependency lockfile.
- Versioned input data, data dictionary, acquisition instructions, and rights notes.
- Calculation traces, source references, assumption records, and exported review artifacts.
- A deterministic baseline and tests for normal, incomplete, contradictory, and failure-recovery cases.
- A recorded complete workflow with a reproducible manifest and explicit limitations.

## Acceptance Criteria
- The report states its sign convention.
- Unit scaling cannot alter the relative gap.
- Do not classify a discrepancy as fraud or a causal demand shock.
- Preserve reported values rather than forcing an identity through fabricated adjustments.

## Contrast Case
Swap which measure is larger. The gap changes sign while its absolute magnitude remains equal.

## Dataset Scope
The attached starter files are prototype inputs. Expected starter answers are visible. They are not independent holdouts.

Full target: Acquire compatible nominal GDP and gross domestic income releases across 40 quarters. Preserve vintage and source-note changes.

Complete target contracts and expert review remain pending. Public observations are not historical release vintages.

## Tools
pandas, statsmodels, Altair

Use an available model appropriate to the task. Record provider, model identifier, version or retrieval date, settings, cost, and latency. Use deterministic code for calculations. Models may extract, retrieve, or explain evidence. They must not invent missing records, override approval rules, or claim professional authority.

## Constraints
Measurement disagreement is not evidence of wrongdoing.

- No fabricated records, source citations, professional endorsements, or expert-adjudication claims.
- No future-data leakage into dated decisions, model fitting, or historical valuation inputs.
- No unauthorized payments, account restrictions, regulatory filings, or production changes.
- No automated appraisal certification, tax-compliance determination, or unsupported fraud accusation.
- No concealed uncertainty or silently invented inputs required for a numerical conclusion.

## Scoring
- Correct domain calculations and justified methods with documented assumptions. Weight: 30.
- Source lineage, temporal integrity, rights checks, and explicit synthetic-data labels. Weight: 25.
- A working application with review, export, error handling, and reproducible state. Weight: 20.
- Adversarial cases, uncertainty, missing-data behavior, and sensitivity analysis. Weight: 15.
- Evidence-linked conclusions that distinguish facts, assumptions, and unresolved judgments. Weight: 10.
Score each dimension from zero to two. Weighted score = sum(weight * score / 2). A failed safety gate blocks acceptance.