Dask-expr

- Fix read_csv with positional usecols (:pr:`1069`) `Patrick Hoefler`_
- Fix isin for head computation (:pr:`1068`) `Patrick Hoefler`_
- Fix isin with strings (:pr:`1067`) `Patrick Hoefler`_
- Use ensure_deterministic kwarg instead of config (:pr:`1064`) `Florian Jetter`_
- Add cache argument to ``lower_once`` (:pr:`1059`) `Richard (Rick) Zamora`_
- Fix non-integer divisions in FusedIO (:pr:`1063`) `Patrick Hoefler`_
- Fix dropna before merge (:pr:`1062`) `Patrick Hoefler`_
- Fix sort_values for unordered categories (:pr:`1058`) `Patrick Hoefler`_
- Fix to_parquet in append mode (:pr:`1057`) `Patrick Hoefler`_

- Add a bunch of docs (:pr:`1051`) `Patrick Hoefler`_
- reduce pickle size of parquet fragments (:pr:`1050`) `Florian Jetter`_
- Generalize ``get_dummies`` (:pr:`1053`) `Richard (Rick) Zamora`_
- Fixup failing test (:pr:`1052`) `Patrick Hoefler`_
- Add support for ``DataFrame.melt`` (:pr:`1049`) `Richard (Rick) Zamora`_
- Fix default name conversion in `ToFrame` (:pr:`1044`) `Richard (Rick) Zamora`_
- Optimize when from-delayed is called (:pr:`1048`) `Patrick Hoefler`_

- Fix delayed in fusing with multipled dependencies (:pr:`1038`) `Patrick Hoefler`_
- Fix ``drop`` with ``set`` (:pr:`1047`) `Patrick Hoefler`_
- Fix ``None`` min/max statistics and missing statistics generally (:pr:`1045`) `Patrick Hoefler`_
- Fix xarray integration with scalar columns (:pr:`1046`) `Patrick Hoefler`_
- Fix ``shape`` returning integer (:pr:`1043`) `Patrick Hoefler`_
- Fix bug in ``Series`` reductions (:pr:`1041`) `Richard (Rick) Zamora`_

- Fix shuffle after ``set_index`` from 1 partition df (:pr:`1040`) `Patrick Hoefler`_
- Fix loc slicing with Datetime Index (:pr:`1039`) `Patrick Hoefler`_
- Fix loc accessing index for element wise op (:pr:`1037`) `Patrick Hoefler`_
- Fix backend dispatching for ``read_csv`` (:pr:`1028`) `Richard (Rick) Zamora`_
- Add cudf support to ``to_datetime`` and ``_maybe_from_pandas`` (:pr:`1035`) `Richard (Rick) Zamora`_

- Move IO docstrings over (:pr:`1033`) `Patrick Hoefler`_
- Fuse more aggressively if parquet files are tiny (:pr:`1029`) `Patrick Hoefler`_
- Add nr of columns to explain output for projection (:pr:`1030`) `Patrick Hoefler`_
- Fix error in analyze for scalar (:pr:`1027`) `Patrick Hoefler`_
- Fix doc build error (:pr:`1026`) `Patrick Hoefler`_
- Add docs for usefule optimizer methods (:pr:`1025`) `Patrick Hoefler`_
- Rename uniuqe_partition_mapping property and add docs (:pr:`1022`) `Patrick Hoefler`_
- Fix read_parquet if directory is empty (:pr:`1023`) `Patrick Hoefler`_
- Fix assign after set index incorrect projections (:pr:`1020`) `Patrick Hoefler`_
- Use implicit knowledge about divisions for efficient grouping (:pr:`946`) `Florian Jetter`_
- Simplify dtype casting logic for shuffle (:pr:`1012`) `Patrick Hoefler`_
- Fix column projections in merge when suffixes are relevant (:pr:`1019`) `Patrick Hoefler`_

- Fix `unique` with numeric columns (:pr:`1017`) `Patrick Hoefler`_
- Fix projection for rename if projection isn't renamed (:pr:`1016`) `Patrick Hoefler`_
- Fix head for npartitions=-1 and optimizer step (:pr:`1014`) `Patrick Hoefler`_
- Deprecate to/from_dask_dataframe API (:pr:`1001`) `Richard (Rick) Zamora`_