force floating point math for shape calculations

[bmcfee]

This PR force-casts sampling rates to be floating point values when calculating the output shape of the resampled array.

This is designed to circumvent numerical overflows that arise when sr_new is a fixed-precision integer (eg np.int16) for some reason. This didn't arise in our tests here, but did pop up over in librosa/librosa#1569

[codecov-commenter]

⚠️ Please install the to ensure uploads and comments are reliably processed by Codecov.

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 75.22%. Comparing base (9c9fa90) to head (69e42e8).
Report is 30 commits behind head on main.

❗ Your organization needs to install the Codecov GitHub app to enable full functionality.

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #115   +/-   ##
=======================================
  Coverage   75.22%   75.22%           
=======================================
  Files           4        4           
  Lines         113      113           
=======================================
  Hits           85       85           
  Misses         28       28           

Flag	Coverage Δ
unittests	75.22% <100.00%> (ø)

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.