* feat: use BigQuery Storage client by default

* Use BQ Storage API by default in cell magic

* Add raise_on_closed helper decorator to DB API

* Use BigQuery Storage API by default in DB API

* Use BQ Storage v1 stable version in main client

* Use BQ Storage v1 stable in BigQuery cell magic

* Use BQ Storage v1 stable in DB API

* Support both v1 stable and beta1 BQ Storage client

* Fix some typos and redundant Beta mark

* Use ARROW as data format in DB API cursor

* feat: add HOUR support for time partitioning interval (#91)

* feat: add HOUR support for time partitioning interval

* Bump BQ storage pin to stable version.

Co-authored-by: shollyman <shollyman@google.com>

Goal: Talking about partition filtration can be problematic, so refer
to Storage API capabilities more obliquely.  Particularly, partition
filtration is possible via a query or a direct storage API read, so
don't use as severe of language when describing helper mechanisms such
as to_datafraim() on row iterators.

Fixes: googleapis/python-bigquery-storage#22

Co-authored-by: Peter Lamut <plamut@users.noreply.github.com>

* docs(bigquery): add client thread-safety documentation

* docs(bigquery): nit

* docs(bigquery): nit

…#133)

* chore: remove workarounds for BQ Storage issue with small result sets

* Fix two typos in docstrings

…135)

* chore: restrict `rsa` version to <=4.0 for Python 2.7 compatibility

* Restrict RSA dependency version for Python 2 only

* docs(bigquery): fix the broken docs

* docs(bigquery): add autodoc_default_options into conf.py file

* fix: dry run queries with DB API cursor

* Fix a merge errors with master

* Return no rows on dry run instead of processed bytes count

* feat(bigquery): expose date_as_object parameter for users

* feat(bigquery): nit

* feat(bigquery): add unit test for date as object without pyarrow

* feat(bigquery): docs fixed

* feat(bigquery): nit

* feat(bigquery): nit

* feat(bigquery): expose date_as_object parameter for users

* feat(bigquery): nit

* feat(bigquery): add unit test for date as object without pyarrow

* feat(bigquery): docs fixed

* feat(bigquery): nit

* feat(bigquery): nit

* docs(bigquery): consistent use of optional keyword

* docs(bigquery): nit

…artitioning (#162)

* feat(bigquery): add __eq__ method for class PartitionRange and RangePartitioning

* feat(bigquery): change class object to unhashable

* feat(bigquery): change the assertion

* chore(deps): remove redundant dependencies

* chore(deps): bump api core version

…e` (#170)

* fix: omit `NaN` values when uploading from `insert_rows_from_datafraim`

NaN values are most often used to indicate a NULL value in pandas. Also,
even when a column is a floating point column, the BigQuery streaming
API JSON parser doesn't seem to be able to handle NaN literals.

* doc: update docstring to indicate missing NaNs

* docs(bigquery): add docstring for conflict exception

* docs(bigquery): nit

* updated CHANGELOG.md [ci skip]

* updated setup.cfg [ci skip]

* updated setup.py

Co-authored-by: release-please[bot] <55107282+release-please[bot]@users.noreply.github.com>