`DataTree.to_dict()` method does not behave as expected

What happened?

I am working with DataTree and find it very useful! However, I think I found a bug in the .to_dict() method.
When I build a DataTree from a dict with DataTree.from_dict() and then want to get the dict again with DataTree.to_dict() the resulting dict differs from the original one.

What did you expect to happen?

I expected the two dicts to be the same. Instead, the root node receives a dict entry with an empty Dataset. I argue that empty nodes should not appear in the dict.

Minimal Complete Verifiable Example

import xarray as xr
from xarray.core.datatree import DataTree

example_dict = {"set1": xr.Dataset({"var1": xr.DataArray([1, 2, 3], dims = "time")}),
                "set2": xr.Dataset({"var1": xr.DataArray([7, 8, 9], dims = "time")})}
dt = DataTree.from_dict(example_dict)
dt.to_dict() == example_dict # False
# should be True imo

# or even just
DataTree.from_dict({}).to_dict == {} # False

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

No response

Anything else we need to know?

No response

Environment

Start with the DataTree.from_dict() and DataTree.to_dict() entry points using the minimal Python example in the issue. Check how the root node and empty Dataset are represented during the round trip. Done means converting the provided dictionaries to a DataTree and back produces the original dictionaries, including an empty input producing {}.