Handle monthly resolution timeseries in upgrades_visualizer

Resolves #[issue number here].

Pull Request Description

Backend Changes:

[x] Breakdown the monolithic upgrades_visualizer.py into separate modules:
- viz_data.py -> This contains the VizData class for handling athena queries and processing data
- figure.py -> This contains the UpgradesPlot class for handling plotly figure generation
- plot_utils.py -> utility functions
- upgrades_visualizer.py -> Houses the ploty dash app with the associated callback for rendering the plot webpage.
[x] Use polars instead of pandas for data processing
[x] Add some tests for the viz_data related portion of the upgrades_visualizer.

UI Changes:

[x] Added a top level "Resolution" option with "annual" and "monthly" options.
[x] Removed "Applied only" check box since it was more confusing than helpful.
[x] Removed "show all points" check box used for box plot since Scatter plot provides similar functionality
[x] Editable tiles - the axis labels and plot titles can be edited. Useful when savings the plot image as png
[x] group_by dropdown now includes "month" and can be used when using monthly resolution

Some screenshots below: