transpose() does not appear to change dataset dimension order.

What happened?

Hello,

I am using .transpose() on an xr.Dataset in order to change the dimension order to make it cf-compliant, e.g. by having the dimension order be TIME, X, Y. While it appears to work when I do a print on the dataset, the dimension order doesn't actually change.

What did you expect to happen?

I expect that the listed dimension order (ds.dims) updates, so that I can specify the exact order of the dimensions as required for CF-compliance.

Minimal Complete Verifiable Example

import xarray as xr
import pandas as pd
import numpy as np

# Dataset construction taken from xr.Dataset documentation
np.random.seed(0)
temperature = 15 + 8 * np.random.randn(2, 3, 4)
precipitation = 10 * np.random.rand(2, 3, 4)
lon = [-99.83, -99.32]
lat = [42.25, 42.21]
instruments = ["manufac1", "manufac2", "manufac3"]
time = pd.date_range("2014-09-06", periods=4)
reference_time = pd.Timestamp("2014-09-05")

ds = xr.Dataset(
    data_vars=dict(
        temperature=(["loc", "instrument", "time"], temperature),
        precipitation=(["loc", "instrument", "time"], precipitation),
    ),
    coords=dict(
        lon=("loc", lon),
        lat=("loc", lat),
        instrument=instruments,
        time=time,
        reference_time=reference_time,
    ),
    attrs=dict(description="Weather related data."),
)

ds_transposed = ds.transpose('time','loc','instrument')

# Order has not changed, returns True
ds_transposed.dims == ds.dims

MVCE confirmation

• Minimal example — the example is as focused as reasonably possible to demonstrate the underlying issue in xarray.
• Complete example — the example is self-contained, including all data and the text of any traceback.
• Verifiable example — the example copy & pastes into an IPython prompt or Binder notebook, returning the result.
• New issue — a search of GitHub Issues suggests this is not a duplicate.
• Recent environment — the issue occurs with the latest version of xarray and its dependencies.

Relevant log output

<frozen _collections_abc>:834: FutureWarning: The return type of `Dataset.dims` will be changed to return a set of dimension names in future, in order to be more consistent with `DataArray.dims`. To access a mapping from dimension names to lengths, please use `Dataset.sizes`.
<frozen _collections_abc>:894: FutureWarning: The return type of `Dataset.dims` will be changed to return a set of dimension names in future, in order to be more consistent with `DataArray.dims`. To access a mapping from dimension names to lengths, please use `Dataset.sizes`.
True

Anything else we need to know?

No response

Environment

Start with the MVCE in the issue and inspect Dataset.transpose() together with Dataset.dims, comparing the returned dataset's dimension-order representation with the expected time, loc, instrument order. Use the example as a regression check and confirm the documented distinction between Dataset.dims and Dataset.sizes before defining done.