Campaign management/Non regression — execution history and release comparison

[ThomasMalletCodra]

Description

Add the ability to track test execution history over time and compare results between releases or branches. This turns ModuleTester from a single-run tool into a continuous quality tracking system.

Current behavior

• Each test execution is independent — no history is stored.
• Previous results are lost when a test is re-run.
• No way to compare results between two versions of a project.

Expected behavior

• Each test plan execution is saved as a "campaign" with a timestamp, project version, and results.
• A campaign history view shows all past executions for a project.
• Side-by-side comparison between two campaigns: new failures, fixed tests, regressions.
• Campaigns can be tagged with a version/branch/commit identifier.
• Export campaign history as a summary report.

Implementation ideas

1. Store campaigns as JSON/SQLite in a .moduletester/campaigns/ directory.
2. Each campaign record: {timestamp, version, git_commit, results: [{test, status, duration, output}]}.
3. Add a "Campaign History" panel in the GUI with a table of past runs.
4. Add a "Compare" action that diffs two selected campaigns.
5. Highlight regressions (was passing, now failing) and fixes (was failing, now passing).

Acceptance criteria

• Test executions are persisted as campaigns
• Campaign history is viewable in the GUI
• Two campaigns can be compared side-by-side
• Regressions and fixes are highlighted
• Campaigns can be tagged with version/commit metadata
• Campaign data can be exported