v0.3.0

What's Changed

• Fix #92: ship benchmark-aware attribution and tearsheets by @seans-openclawbot in #96
• Fix #90: add target-weight portfolio construction workflow by @seans-openclawbot in #99
• chore(deps): bump rand from 0.8.5 to 0.9.3 by @dependabot[bot] in #98
• Fix #89: ship real optimization workflow by @seans-openclawbot in #94
• Fix #91: add a durable experiment registry by @seans-openclawbot in #95
• Fix #88: replace mock backtests with the real engine by @seans-openclawbot in #93

Full Changelog: v0.2.0...v0.3.0