[Bug]: xarray=2022.11.0 .plot() breaks with time coordinates represented by cftime

[tomvothecoder]

What happened?

xarray's .plot wrapper requires the nc-time-axis package to plot time coordinates represented by cftime objects.

If nc-time-axis does not exist, xarray throws the error:
ImportError: Plotting of arrays of cftime.datetime objects or arrays indexed by cftime.datetime objects requires the optional nc-time-axis (v1.2.0 or later) package.

UPDATE: It looks like installing nc-time-axis does not fix the problem. It produces this error now: TypeError: float() argument must be a string or a real number, not 'cftime._cftime.DatetimeProlepticGregorian'

I traced the error back to xarray=2022.11.0: pydata/xarray#7275. The nc_time_axis import in xarray was accidentally removed and has since been restored in subsequent versions.

What did you expect to happen?

Plotting should work fine like in previous versions of xarray.

Minimal Complete Verifiable Example

import xcdat as xc

filepath = "https://esgf-data1.llnl.gov/thredds/dodsC/css03_data/CMIP6/CMIP/CSIRO/ACCESS-ESM1-5/historical/r10i1p1f1/Amon/tas/gn/v20200605/tas_Amon_ACCESS-ESM1-5_historical_r10i1p1f1_gn_185001-201412.nc"

ds = xc.open_dataset(
    filepath,
    add_bounds=True,
    decode_times=True,
    center_times=True
)

# Unit adjustment from Kelvin to Celcius.
ds["tas"] = ds.tas - 273.15

ds_trop_avg = ds.spatial.average("tas", axis=["X","Y"], lat_bounds=(-25,25))

ds_trop_avg.tas.isel(time=slice(0, 100)).plot()

Relevant log output

No response

Anything else we need to know?

We need to update the xarray version in our conda environment yml files to the latest one.

Environment

xcdat=0.4.0