Added DPZ extractor to pargen

[SamuelBorden]

The function get_dpz_decay_constants works to get the constants needed for the double pole zero filter in three steps:

1. A guess for the long and short time constants are obtained by taking linear fits to the logarithm of a superpulse
2. These guesses are used to seed a least squares fit
3. The least squares values are then used in a dsp chain that optimizes the flatness of the decaying tails of waveforms

This is a first pass at the routine, so we should think maybe a bit more about where to put this in the dataflow. Right now this quickly interfaces with the single pole-zero dataflow script; to do that, the superpulse is created inside the function get_dpz_decay_constants. But we might want to make get_dpz_decay_constants a bit more general and instead create the superpulse in the dataflow.

There are other issues currently with the double pole zero filter, see https://github.com/legend-exp/dspeed/pull/52. So I am going to mark this as a draft PR

[codecov[bot]]

Codecov Report

Attention: Patch coverage is 62.98343% with 67 lines in your changes are missing coverage. Please review.

Project coverage is 25.51%. Comparing base (ae3752a) to head (7fdd904). Report is 50 commits behind head on main.

Files	Patch %	Lines
src/pygama/pargen/extract_tau.py	62.98%	67 Missing :warning:

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #558      +/-   ##
==========================================
+ Coverage   22.82%   25.51%   +2.68%     
==========================================
  Files          42       42              
  Lines        8110     8220     +110     
==========================================
+ Hits         1851     2097     +246     
+ Misses       6259     6123     -136     

:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.

[gipert]

@ggmarshall @SamuelBorden can this be merged to main or will mess up with the refactor?

[SamuelBorden]

I need to change some imports from the energy cal routines, but then it should be good to go. I'll work on synergizing this with the pargen refactor