Dataiter

=========================

* `Vector`: Handle timedeltas correctly for NA checks and printing
* `Vector.is_timedelta`: New method

=========================

* `DataFrame.sort`: Convert object to string for sorting
* `Vector.sort`: Convert object to string for sorting
* Fix conditional Numba use when importing the numba package works,
but caching doesn't
* Add `di-open` cli command (currently not part of the default install,
but can be installed from source using `make install-cli`)

=========================

* `DataFrame.modify`: Add support for grouped modification (19)
* `DataFrame.split`: New method
* `ListOfDicts.split`: New method

=========================

* `DataFrame.compare`: New experimental method
* `Vector.as_string`: Add `length` argument
* Change the documentation to default to the latest release ("stable")
instead of the development version ("latest")

=========================

* Use keyword-only arguments where appropriate – the general principle
is that mandatory arguments are allowed as positional, but optional
modifiers are keyword only
* Rename all instances of "missing" to "na", such as `Vector.is_missing`
to `Vector.is_na`, the only exception being
`ListOfDicts.fill_missing`, which becomes
`ListOfDicts.fill_missing_keys`
* Truncate data frame object and string columns at
`PRINT_TRUNCATE_WIDTH` (default 32) for printing

===========================

* Fix aggregation functions to work with all main data types:
boolean, integer, float, date, datetime and string
* Fix aggregation functions to handle all missing values (NaN, NaT,
blank string) correctly, the same as implemented in Vector
* Rename aggregation functions' `dropna` arguments to `drop_missing`
* `first`, `last`, `nth`: Add `drop_missing` argument
* `Vector.drop_missing`: New method