Pipelinewise-target-snowflake

-------------------
- Add parquet support
- Add check and few logs in the date parsing routine
- Bumping dependencies

-------------------
- Update caching mechanism to fix issue with badly ordered queryies in a transaction
- Introduced a reserved named parameter for prepared statements.
- Do not use parallel file upload with PUT command and table stages.
- Bumping dependencies

-------------------

- Add `{{database}}` token to `query_tag` parameter
- Use Jinja style `query_tag` template variables

-------------------

- Fixed a dependency issue

-----------------------------------------

- Use snowflake table stages by default to load data into tables
- Add optional `query_tag` parameter
- Add optional `role` parameter to use custom roles
- Fixed an issue when generated file names were not compatible with windows
- Bump `joblib` to `0.16.0` to be python 3.8 compatible
- Bump `snowflake-connectory-python` to `2.3.6`
- Bump `boto3` to `1.16.20`

-------------------

- Fixed an issue when `pipelinewise-target-snowflake` failed when `QUOTED_IDENTIFIERS_IGNORE_CASE` snowflake parameter set to true
- Add `aws_profile` option to support Profile based authentication to S3
- Add option to authenticate to S3 using `AWS_PROFILE`, `AWS_ACCESS_KEY_ID`, `AWS_SECRET_ACCESS_KEY` and `AWS_SESSION_TOKEN` environment variables
- Add `s3_endpoint_url` and `s3_region_name` options to support non-native S3 accounts
- Flush stream only if the new schema is not the same as the previous one