Add more add_location tests

[emiliom]

Addresses all but one of the tests specified in #769.

test_consolidate_integration.py::test_add_location now includes two additional sample raw files, including an AZFP file. The test files now encompass:

• EK60 with Platform lat-lon data (the file previously tested)
• EK60 with no lat-lon data (from OOI)
• AZFP with no lat-lon data, where the test uses update_location to first add a fixed-location single point

The tests in this PR currently include:

• Previously existing, but now run also on an AZFP file:
  • Presence of latitude and longitude variables
  • Absence of time variables
  • Running the above set of tests on nmea_sentence="GGA" for the EK60 dataset with Platform lat-lon data
• New tests:
  • When source echodata object has no valid lat-lon data, test for expected ValueError
  • No longitude or latitude value is null
  • longitude and latitude variables have a single dimension, ping_time
  • (Added in latest commit) Interpolated lat/lon values added by add_latlon function match up with values interpolated in the test
  • (Added in latest commit) For single-point case, lat-lon coordinates each contains a single unique value that matches the value passed to update_platform

Remaining test: Checking the added (interpolated) lat/lon values between the add_latlon function and directly computed from scipy.interpolate

Also, see #769 (comment) for findings about time1 vs time3 dimensions associated with water_level, and for the outcome of the first interpolated latitude and longitude value being nan. In this PR I modified add_location to set the first value of interpolated latitude and longitude equal to the second value use the scipy.interpolate.interp1d fill_value="extrapolate" scheme. Let's discuss this.

[codecov-commenter]

Codecov Report

Merging #1000 (5b0e1c0) into dev (b802e8c) will increase coverage by 12.47%.
The diff coverage is 100.00%.

📣 This organization is not using Codecov’s GitHub App Integration. We recommend you install it so Codecov can continue to function properly for your repositories. Learn more

@@             Coverage Diff             @@
##              dev    #1000       +/-   ##
===========================================
+ Coverage   79.63%   92.10%   +12.47%     
===========================================
  Files          66        3       -63     
  Lines        5745      152     -5593     
===========================================
- Hits         4575      140     -4435     
+ Misses       1170       12     -1158     

Flag	Coverage Δ
unittests	92.10% <100.00%> (+12.47%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
echopype/consolidate/api.py	94.52% <100.00%> (+2.73%)	⬆️

... and 63 files with indirect coverage changes

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more

[emiliom]

Ok, done. I decided to use xarray interpolate in the test rather than scipy.interpolate, for expedience. I don't think we gain much by using scipy.interpolate; or at least, I don't think it's worth the extra effort right now.

[leewujung]

These will have to change if we do not extrapolate the first position.

[emiliom]

I've removed these tests. Now that nan lon & lat values may be valid, I can't think of a general, useful test involving nan's. But I think the other tests already cover all ground: lon & lat contain the ping_time dimension (and no other dim), and the values match against interpolated values (that may include nan's) or fixed-point broadcasting. BTW, I modified the comparison of interpolated values to account for nan's.

[leewujung]

@emiliom : I am done reviewing this. My comments are mostly small and some completely textual. The only exception is the one involving extrapolation. Let me know what you think on that.

[emiliom]

Thanks! I'll remove the extrapolation and update the tests to reflect that change.

[leewujung]

Looks good. Thanks @emiliom. Please feel free to merge this!