Fixes for numpy 2.0

[samuelbray32]

numpy 2.0 introduced new errors. Required changes:

• columns object returned by np.fromfile now have a second axis of one dimesnion

  • Source of error:

  np.dtype(("f8", 1) will now return a shape 1 subarray dtype rather than a non-subarray one.

  • This raised errors when converting to pandas dataframe
  • Add function convert_datafile_to_pandas to do this cleanly

• Numpy changed promotion rules of datatypes in 2.0

  • memap now required explicit casting to int16 in analog signal processing in SpikeGadgetsRawIO

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 90.34%. Comparing base (fbf92a8) to head (f34e7ba).
Report is 5 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #104      +/-   ##
==========================================
+ Coverage   90.32%   90.34%   +0.01%     
==========================================
  Files          25       25              
  Lines        2429     2433       +4     
==========================================
+ Hits         2194     2198       +4     
  Misses        235      235              

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