Adding TPF translator and parser

[bmorris3]

Try it out with:

import astropy.units as u
from astropy.time import Time
from lightkurve import search_lightcurve, search_targetpixelfile

from lcviz import LCviz

# download a Kepler TPF:
tpf = search_targetpixelfile('KIC 6922244', author="Kepler", quarter=9, cadence="long").download()

# download a TESS TPF:
# tpf = search_targetpixelfile('TIC 307210830', mission="TESS", cadence="long").download()


# Load the light curve into LCviz:
lcviz = LCviz()
lcviz.load_data(tpf)
lcviz.show()

[codecov]

Codecov Report

Attention: 25 lines in your changes are missing coverage. Please review.

Comparison is base (8f5d02e) 93.91% compared to head (e129bef) 93.30%.

Files	Patch %	Lines
lcviz/utils.py	83.66%	25 Missing ⚠️

Additional details and impacted files

@@               Coverage Diff               @@
##           feature-tpf      #75      +/-   ##
===============================================
- Coverage        93.91%   93.30%   -0.61%     
===============================================
  Files               35       35              
  Lines             1545     1689     +144     
===============================================
+ Hits              1451     1576     +125     
- Misses              94      113      +19     

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

[kecnry]

Oh I'm a bit surprised to see this actually plot the light curve... where is that happening? Do we want that? Or do we want to force the logic through the (eventual) extraction plugin?

[kecnry]

TODO for this PR:

• generalize to all TPF files or at least Kepler, K2, Tess
• test coverage
• decide about cube shape/transpose (see Adding TPF translator and parser #75 (comment))

Follow-up efforts (out-of-scope here):

• filter/exclude TPF entries from lc viewer data menus (might need logic or refactoring upstream)
• plugin-managed viewer to visualize cube (with appropriate data filtering in menu)
• extraction plugin with live-preview and pipeline defaults
• slice plugin with indicator in all time viewers
• mouseover support for cube viewer(s)

[kecnry]

It seems that the resulting cube is transposed with respect to the convention in cubeviz. Should we handle this here or elsewhere?

To test, try the following with my tpf-viewer branch, built on top of this:

from lightkurve import search_targetpixelfile
from lcviz import LCviz

tpf = search_targetpixelfile("HAT-P-11", mission="Kepler", cadence="long", quarter=10).download()

lcviz = LCviz()
lcviz.load_data(tpf)
lcviz.show()

lcviz.get_data().shape
# (4447, 22, 11) which is (time, x, y) whereas cubeviz uses (x, y, spectral_axis)

[kecnry]

Test coverage should fill in as this actually becomes usable 🤞