v0.4.0

Full Changelog: v0.3.0...v0.4.0

v0.3.0

v0.3.0

This release makes the performance evidence and observability paths operational end to end.

Highlights

• OTEL metrics are exported to Prometheus and available through a provisioned Grafana datasource.
• Main load runs enforce the committed measured baseline, with positive and negative CI fixtures.
• Spike and breakpoint thresholds are centralized; stress and soak workloads enforce endpoint SLOs.
• Human-readable summaries include business metrics and the report site charts latency trends.
• Observability health checks work on Windows, Linux, and macOS.
• Concurrent-reader behavior, soak analysis, and authenticated onboarding are now explicit.

v0.2.0

Highlights

• Failure-sensitive performance evidence and regression handling
• Stronger workload modeling, capacity gates, and segmented execution
• Improved configuration validation, observability, reporting, and contributor workflows
• Repository-owned secret scanning without licensed CI dependencies
• Stable regression execution under negligible hosted-runner scheduling jitter

Included commits

• f121086 fix: make performance evidence failure-sensitive
• 247bd0d feat: strengthen workload modeling and performance reporting
• cb7796f feat: harden configuration observability and contributor workflows
• fb281b7 feat: add capacity gates and segmented execution
• 2b00d86 fix: remove licensed secret scan dependency
• 82ecfb2 fix: tolerate negligible load scheduler jitter
• 4d071cc fix: handle pending performance baseline

All changes are included through merge commit acdb529.