Alignment: allow flexible index coordinate order

[benbovy]

• Closes Custom indexes and coordinate (re)ordering #7002
• Tests added
• User visible changes (including notable bug fixes) are documented in whats-new.rst

This PR relaxes some of the rules used in alignment for finding the indexes to compare or join together. Those indexes must still be of the same type and must relate to the same set of coordinates (and dimensions), but the order of coordinates is now ignored.

It is up to the index to implement the equal / join logic if it needs to care about that order.

Regarding pandas.MultiIndex, it seems that the level names are ignored when comparing indexes:

midx = pd.MultiIndex.from_product([["a", "b"], [0, 1]], names=("one", "two")))
midx2 = pd.MultiIndex.from_product([["a", "b"], [0, 1]], names=("two", "one"))

midx.equals(midx2)  # True

However, in Xarray the names of the multi-index levels (and their order) matter since each level has its own xarray coordinate. In this PR, PandasMultiIndex.equals() and PandasMultiIndex.join() thus check that the level names match.

[benbovy]

Considering

midx = pd.MultiIndex.from_product([["a", "b"], [0, 1]], names=("one", "two"))
midx_coords = xr.Coordinates.from_pandas_multiindex(midx, "x")
ds = xr.Dataset(coords=midx_coords)
# <xarray.Dataset>
# Dimensions:  (x: 4)
# Coordinates:
#   * x        (x) object MultiIndex
#   * one      (x) object 'a' 'a' 'b' 'b'
#   * two      (x) int64 0 1 0 1
# Data variables:
#     *empty*

midx2 = pd.MultiIndex.from_product([["a", "b"], [0, 1]], names=("two", "one"))
midx2_coords = xr.Coordinates.from_pandas_multiindex(midx2, "x")
ds2 = xr.Dataset(coords=midx2_coords)
# <xarray.Dataset>
# Dimensions:  (x: 4)
# Coordinates:
#   * x        (x) object MultiIndex
#   * two      (x) object 'a' 'a' 'b' 'b'
#   * one      (x) int64 0 1 0 1
# Data variables:
#     *empty*

Still with this PR, we have this:

ds.equals(ds2)   # True

Which doesn't make sense. This is because Dataset.equals() doesn't make any call to Index.equals() and still relies here on pd.MultiIndex.equals(). We'll need to change that.

[benbovy]

Hmm pd.MultiIndex.equals ignoring the level names seems to be a bug in pandas actually: pandas-dev/pandas#32569.

[dcherian]

It is up to the index to implement the equal / join logic if it needs to care about that order.

I think this is the right decision.

ping @shoyer for input