[FIX] ASR: init clean_windows rejection masks before one-sided zthresholds

[sappelhoff]

In clean_windows, mask1 and mask2 were only assigned inside their respective if np.max(zthresholds) > 0 / if np.min(zthresholds) < 0 branches, but combined unconditionally via np.logical_or.reduce((mask1, mask2, mask3)). Since zthresholds is a documented, user-facing parameter, passing a one-sided value (e.g. [1, 2] or [-5, -1]) skips one branch and raises UnboundLocalError. The default [-3.5, 5] straddles 0 so both branches run, keeping the bug latent.

Initialize both masks to all-False before the conditionals, so a criterion that isn't evaluated simply rejects no windows.

Testing

Added a regression test exercising clean_windows with both one-sided directions of zthresholds. Full tests/test_asr.py passes; ruff clean.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 83.11%. Comparing base (6b107c9) to head (c2b9ca1).

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #113      +/-   ##
==========================================
+ Coverage   83.10%   83.11%   +0.01%     
==========================================
  Files          25       25              
  Lines        2823     2825       +2     
==========================================
+ Hits         2346     2348       +2     
  Misses        477      477              

☔ View full report in Codecov by Harness.
📢 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.

[sappelhoff]

Note: includes the nonlinear_eigenspace reproducibility fix from #118. test_asr_class[*-riemann] compares incremental-vs-bulk output with a loose tolerance and is flaky on CI because that path used an unseeded RNG; this can trip on any PR regardless of what it changes. #118 is the root fix — once it merges first, this commit drops out on rebase.