SNOW-649339: Allow usage of a pre-configured NAMED File Format working with data in staged files

[brian-kalinowski-sonos]

What is the current behavior?

Currently all the DataFrameReader file methods (.csv, .parquet, .json, etc) create temp file formats with options from python dicitionaries.

Even if the external stage has a default (name) file format attached the reader methods can fail when/if file format options are not passed to the DataFrameReader.

This also ties into the copy_into_table method.

What is the desired behavior?

Be able to pass a pre-configured file format name and/or have the DataFrameReader methods use the attached file format to the external stage.

How would this improve snowflake-snowpark-python?

• Allow users not to have to re-configure file format options in snowpark, just use the file format/stage that's already set up in snowflake.

• Mix and match different file formats between stages without having to reset all the file/copy options again.

References, Other Background

https://github.com/snowflakedb/snowpark-python/blob/main/src/snowflake/snowpark/_internal/analyzer/snowflake_plan.py#L685 Could bypass this query if a user passes a named file format, or the option to use the file format configured for the selected stage.

[sfc-gh-jdu]

We will support the file format name in copy_into_table method soon. cc @sfc-gh-mkeller

[sfc-gh-jdu]

https://github.com/snowflakedb/snowpark-python/pull/455 addressed this issue