Draft: month-cell: Auto-expand month names on 1st day of the month when not conflicting with weather forecast

This is the experimental follow-up that builds on top of !478

It currently uses a simple heuristic/assumption for when it is safe to show the long version of the month name: when there is no weather forecast label competing for the same days in the cell of the 1st day of the month.

I'm wondering if this approach would be acceptable. If so, I may need a sanity check from @camelCaseNick too, and a trick to either get the GCAL_WEATHER_FORECAST_MAX_DAYS_DEFAULT value dynamically (instead of hardcoding 5 as the number of days for weather forecast danger zone), or perhaps an easy way to determine if the current cell is a cell with weather_info in it (I have a hunch this might be cleaner, but don't know how to do that check properly).

Update: it now checks whether there is a weather forecast in the cell, rather than just guessing with the number of days distance from today.