Introduce experimental BasisCommonGP2 and FourierBasisCommonGP2 that accept selections

[vallis]

For instance, to implement independent correlated GPs on each telescope, one could do

rn = gp_signals.FourierBasisCommonGP2(pl, orf, selection=Selection(selections.by_telescope), Tspan=Tspan)

Note that it is currently necessary to set Tspan.

The PR also adds the method telescope() to pulsar, and by_telescope() to selections.

[codecov[bot]]

Codecov Report

Merging #290 (72360db) into master (f2b00be) will increase coverage by 0.35%. The diff coverage is 96.87%.

@@            Coverage Diff             @@
##           master     #290      +/-   ##
==========================================
+ Coverage   88.05%   88.41%   +0.35%     
==========================================
  Files          13       13              
  Lines        2980     3055      +75     
==========================================
+ Hits         2624     2701      +77     
+ Misses        356      354       -2     

Impacted Files	Coverage Δ
enterprise/signals/gp_signals.py	91.23% <96.34%> (+0.92%)	:arrow_up:
enterprise/signals/signal_base.py	90.65% <100.00%> (+0.54%)	:arrow_up:

---

Continue to review full report at Codecov.

Legend - Click here to learn more Δ = absolute <relative> (impact), ø = not affected, ? = missing data Powered by Codecov. Last update f2b00be...72360db. Read the comment docs.

[vhaasteren]

Hi @vallis , is this PR ready for review? I can properly look at it this week.

[vhaasteren]

One question: is this a separate class instead of just updating BasisCommonGP, because this is experimental? Ideally the selections would just work for the regular common Common GP classes