Context

Part of #964. Derived from the 2026-06-13 analysis + adversarial plan review. network is honestly pre-1.0 (package 0.1.1) and is the ecosystem long pole: develop is 112 commits ahead of main, and the open XL coverage epic #953 (40->80%) currently appears to gate the tag.

Key finding

coverage.yml:156 runs ctest ... || true with fail_ci_if_error:false and no --fail-under, so #953 does NOT mechanically gate the tag. Keeping the +40-point climb on the critical path puts the slowest item ahead of the SaMD apex (pacs) for no mechanical reason.

Scope / Acceptance Criteria

• Ship network v1.0.0 at the current MEASURED coverage; record the definitive lcov number as the CHANGELOG SSOT (correct the README "~80%" claim to the measured value)
• Demote Expand unit test coverage from 40% to 80% #953 to a v1.1.0 fast-follow epic
• Replace the climb with an ENFORCED coverage-regression floor at the current % (remove || true, set fail_ci_if_error:true) — no backslide
• Reconcile the dual/out-of-order CHANGELOG (future-dated v2.0.0/v1.5.0 entries above "v1.0.0 Pending" while CMake is 0.1.1) into a single monotonic history before the 112-commit release-merge
• Bump 0.1.1 -> 1.0.0 (CMakeLists.txt:50 + both vcpkg.json), promote CHANGELOG [Unreleased] -> [v1.0.0]
• Preconditions: common/logger/monitoring v1.0 tags landed (network CHANGELOG subordinates its tag to the upstream Tier 0-3 epics)