Trino

* Implement `get_catalog_names` for SQLAlchemy to get the list of catalogs
present on the cluster.
([401](https://github.com/trinodb/trino-python-client/issues/401))
* Use `OAuth2Authentication` if `externalAuthentication` parameter is set on
the connection url when using SQLAlchemy.
([343](https://github.com/trinodb/trino-python-client/issues/343))
* Set the `User-Agent` HTTP header in all requests to `Trino Python
Client/<client_version>`.
([411](https://github.com/trinodb/trino-python-client/issues/411))
* Raise `TrinoConnectionError` for all connection related errors.
([364](https://github.com/trinodb/trino-python-client/issues/364))

* Fix SQLAlchemy URL failing to parse if `legacy_prepared_statements` parameter
was specified.
([380](https://github.com/trinodb/trino-python-client/issues/380))
* Fix more than required `EXECUTE IMMEDIATE` queries being executed when
`legacy_prepared_statements` parameter wasn't explicitly set.
([380](https://github.com/trinodb/trino-python-client/issues/380))

* Queries with parameters now uses `EXECUTE IMMEDIATE` instead of `PREPARE`,
`EXECUTE` and `DEALLOCATE` if the server supports `EXECUTE IMMEDIATE`. If
`EXECUTE IMMEDIATE` is not support the client falls back to using `PREPARE`,
`EXECUTE` and `DEALLOCATE`. If you want to always use `PREPARE`, `EXECUTE`
and `DEALLOCATE` set the kwarg `legacy_prepared_statements` to `True`.
([375](https://github.com/trinodb/trino-python-client/issues/375))

* Reduce the number of times `SELECT version()` query is sent to Trino when
using SQLAlchemy.
([371](https://github.com/trinodb/trino-python-client/issues/371))
* Fix `TIMESTAMP WITH TIME ZONE` being mapped to incorrect zones.
([366](https://github.com/trinodb/trino-python-client/issues/366))

* Fix handling of `Decimal` params which use scientific notation.
([347](https://github.com/trinodb/trino-python-client/issues/347))
* Map Trino `UUID` type to Python `uuid.UUID`.
([354](https://github.com/trinodb/trino-python-client/issues/354))
* Map `sqlalchemy.Uuid` type to Trino `UUID`.
([359](https://github.com/trinodb/trino-python-client/issues/359))
* Support using timezone aware `datetime.time` params in prepared statements.
([360](https://github.com/trinodb/trino-python-client/issues/360))
* Allow accessing `ROW` type fields using their names in addition to their
indices.
([338](https://github.com/trinodb/trino-python-client/issues/338))
* Interpret `roles` without catalog name as system roles for convenience.
([341](https://github.com/trinodb/trino-python-client/issues/341))

* Return `rowcount` property on the DB-API `Cursor` when available.
([325](https://github.com/trinodb/trino-python-client/issues/325))
* Provide a read-only property `query` on the DB-API `Cursor` which contains
the query text submitted through the cursor when available.
([323](https://github.com/trinodb/trino-python-client/issues/323))
* Automatically determine `http_scheme`, `host` and `port` from `host` if it's
a valid URL.
([340](https://github.com/trinodb/trino-python-client/issues/340))
* Fix query cancellation to follow the protocol. Before this change cancelling
a query could lead to errors from the Trino server.
([328](https://github.com/trinodb/trino-python-client/issues/328))