makeweight: always use geometric mean for default gain_mid

[baggepinnen]

Summary

• makeweight(low, mid::Number, high) previously fell back to a hard-coded gain_mid = 1 whenever high >= 1. That violates the downstream formula's low < mag < high precondition whenever both low > 1 and high > 1 (e.g. makeweight(2, 1, 3)), silently producing a weight with complex/NaN poles.
• Conservative fix: use sqrt(low*high) only when low and high lie on the same side of 1 — the previously-broken case. When low and high straddle 1 (the typical case) keep the historical default gain_mid = 1. This preserves user-facing behavior for cases like makeweight(1e-3, 10, 10) (low<1<high) which downstream H∞/LQG examples rely on.

Test plan

• Updated existing test for makeweight(0.1, 1, 2) to assert the mid-frequency magnitude is 1 (straddle case, back-compat).
• New regression test: makeweight(2, 1, 3) (both > 1) produces real coefficients and the mid magnitude is sqrt(2*3).
• New symmetric test: makeweight(0.5, 1, 0.1) (both < 1) also uses sqrt(low*high).
• test_manual_hinf.jl and test_weights.jl pass locally with the change.

🤖 Generated with Claude Code

[codecov-commenter]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 91.77%. Comparing base (14a7deb) to head (e6a2ac3).
⚠️ Report is 4 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #147      +/-   ##
==========================================
- Coverage   91.84%   91.77%   -0.07%     
==========================================
  Files          20       20              
  Lines        3064     3065       +1     
==========================================
- Hits         2814     2813       -1     
- Misses        250      252       +2     

Flag	Coverage Δ
unittests	91.77% <100.00%> (-0.07%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.