Potential bug in interp

What happened?

There seems to be an issue with dtypes when calling interp on dask-backed data arrays.

Example:

da = xr.DataArray(
    data=dask.array.array([1, 2, 3], dtype=np.float32),
    dims=("x"),
    coords={"x": np.array([0, 1, 2], dtype=np.float32)},
)
da = da.interp(x=np.array([0.75], dtype=np.float64))
print(da.dtype, da.compute().dtype, da.astype(np.float32).compute().dtype)

this prints out:

float32 float64 float64

so, not only is there a difference before/after compute(), explicitly casting to float32 doesn't seem to have an effect.

What did you expect to happen?

No response

Minimal Complete Verifiable Example

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

Anything else we need to know?

No response

Environment

Start by running the provided interp example with the reported versions and compare the dtype before and after compute(), including the explicit astype call. Trace the interp entry point and its dask-backed execution path. The issue does not state the expected dtype, so confirm the intended behavior before adding a regression test.