Add line info widget to widgets subpackage

[jaladh-singhal]

This PR aims to put line info widget which was being prepared in a notebook (#1187) as a module within the widgets subpackage so that it can be generated easily by constructing the widget object.

The main changes made are:

• Edited analysis.py to make it work with line inf widget:

  • Classmethod from_model did not take packet_filter_mode as argument but it is required for initializing LastLineInteraction (as __init__() takes it), so added it
  • Added filter mode packet_in_nu and renamed packet_nu to packet_out_nu since it only concerned with output_nu
  • output_nu and last_interaction_in_nu didn't always had "Hz" unit, so made sure that they are converted to Quantity 1st

• Added line_info.py module with the class to create line info widget

• Renamed base.py which contained only shell info widget code to shell_info.py and similarly done this renaming in tests and __init__.py

• Added util.py as a helper functions module to current widgets modules

  • create_table_widget was common in both line_info.py and shell_info.py, so removed it from shell_info.py and moved it there
  • Added a TableSummaryLabel class for creating summary label targetting a table widget (required for handling creation/updation of Total Packets label in last line counts table)

• Added test_line_info.py for testing line info module. Note, there's no test file for util.py because it doesn't have any back-end logic but only front-end logic (of creating widgets) and since line_info module uses it so it's already covered.

To-Dos

• Improvement in line info widget as described here
• Add all the functions from notebook and make the OO interface work
• Add docstrings in module
• Create unit-tests
• Add documentation (implemented in Add widgets documentation to tardis docs #1280 since it makes changes to entire widgets docs not only line info)

[codecov]

Codecov Report

Merging #1264 into master will increase coverage by 1.20%.
The diff coverage is 95.94%.

@@            Coverage Diff             @@
##           master    #1264      +/-   ##
==========================================
+ Coverage   80.54%   81.75%   +1.20%     
==========================================
  Files          41       44       +3     
  Lines        3423     3699     +276     
==========================================
+ Hits         2757     3024     +267     
- Misses        666      675       +9     

Impacted Files	Coverage Δ
tardis/widgets/util.py	85.36% <85.36%> (ø)
tardis/widgets/line_info.py	96.15% <96.15%> (ø)
tardis/widgets/shell_info.py	96.00% <100.00%> (ø)
tardis/widgets/tests/test_line_info.py	100.00% <100.00%> (ø)
tardis/widgets/tests/test_shell_info.py	100.00% <100.00%> (ø)

---

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 862025e...b54358b. Read the comment docs.

[epassaro]

Noticed there are some Pandas ix method in analysis.py . This method is already deprecated and should be replaced by loc or iloc.

[jaladh-singhal]

Noticed there are some Pandas ix method in analysis.py . This method is already deprecated and should be replaced by loc or iloc.

@epassaro Yes that file is very old, I'm actually using only very few methods of LastLineInteraction class from analysis.py in line_info.py and none of them are using ix. Now that you have mentioned it, perhaps we can scrap this file altogether and mimic only the functionality I require from it directly within the line_info.py!

I would like to know opinions of reviewers what do they think, should we do this change right here in this PR or get this merged first and then do the changes (to avoid unexpected surprises 😀 )?

[MarkMageeAstro]

This is a pretty minor comment that likely doesn't affect many people, but I've updated my notebooks to have a dark background. This makes it difficult to read some of the text, as it's black on a dark background. This is shown in the attached image. Maybe adding a white background or changing the colour to something like red would fix this?

[MarkMageeAstro]

Some tests are failing. Looks like it's related to the integrated spectrum. Some other recent PRs aren't failing, so could need rebasing?

[MarkMageeAstro]

Some comments left for minor changes. Mostly about adding comments, etc. The comment I left about dark backgrounds shouldn't hold this PR back from being merged.

[jamesgillanders]

See comments above from myself and others - if these are addressed then I am happy to merge

[ycamacho]

This is a great PR @jaladh-singhal! We have finished our review of it and have asked for minor changes. Please look them over - if anything is taking too long to figure out we can make it an issue to fix for later. This PR is nearly ready to be merged!

[ycamacho]

Looks ready to merge to me!

[MarkMageeAstro]

Also looks good to me