Utils Methods and TIE Speedups

[RHammond2]

This PR address the remaining, easy-to-address slowness that bogs down the TIE analysis method with Example 3 run time decreasing to ~2.5-3.5 minutes from the previous ~16 minute runtime. The speedups come from:

1. converting operations relying Pandas Series methods to the underlying NumPy methods and converting the data to NumPy arrays
2. Converting openoa.utils.filters.bin_filter to array math from nested for loops on top of the improvements from 1

One of the tests was changed because the input data was using mixed data types, but all data being passed through the the various utils methods will be converted to their correct data types, or should by the time they're used in an analysis. In either case mixing string and float data made little sense to test the robustness of a method operating on floats.

[ejsimley]

Hi @RHammond2, thanks for making these improvements. I ran the example 03 notebook and the speedup from the new filter implementation is huge! My only request is to run the last few cells of the example 03 notebook so the outputs are already printed when viewing the notebook.

[codecov-commenter]

Codecov Report

Patch coverage: 96.66% and project coverage change: +0.01% 🎉

Comparison is base (aad3ede) 63.30% compared to head (6f08565) 63.32%.

Additional details and impacted files

@@              Coverage Diff               @@
##           develop_v3     #250      +/-   ##
==============================================
+ Coverage       63.30%   63.32%   +0.01%     
==============================================
  Files              28       28              
  Lines            3911     3910       -1     
==============================================
  Hits             2476     2476              
+ Misses           1435     1434       -1     

Files Changed	Coverage Δ
openoa/utils/filters.py	94.84% <95.83%> (+1.29%)	⬆️
openoa/analysis/turbine_long_term_gross_energy.py	77.23% <100.00%> (-0.50%)	⬇️
openoa/utils/timeseries.py	86.95% <100.00%> (ø)

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