SNOW-3264599: Python - implement cursor reset method

[sfc-gh-asolarski]

TL;DR

Implemented proper cursor state management by adding a reset() method and updating close() to properly clean up resources and return status indicators.

What changed?

• Added reset() method: Clears all result-related state including execute_result, _iterator, _description, _sqlstate, _fetch_mode, _binding_data, and _rownumber. The closing parameter preserves _rowcount when set to True.
• Enhanced close() method: Now calls reset(closing=True), clears the messages list, and returns a boolean indicating success (True), already closed (False), or exception (None).
• Updated execute() method: Added _do_reset parameter (defaults to True) and moved the reset logic to call reset() at the beginning instead of manually clearing state afterward.
• Modified executemany() method: Now calls reset() once before the loop for client-side binding scenarios and passes _do_reset=False to individual execute() calls to avoid redundant resets.

Why make this change?

This change provides proper cursor lifecycle management by centralizing state cleanup logic in the reset() method. It ensures consistent behavior across different execution patterns, prevents state leakage between queries, and provides better error handling and status reporting for cursor operations. The close() method now properly indicates its success status, and executemany() avoids unnecessary resets between individual executions while maintaining clean state.

[sfc-gh-asolarski]

• SNOW-3264599: core + python - abort query method #750
• SNOW-3264599: core + python - query_result fetching by sfqid #747
• SNOW-3264599: Python - describe method #735
• SNOW-3264599: Python - implement cursor reset method #723 👈 (View in Graphite)
• SNOW-3243341: python wrapper - support pandas and pyarrow fetching #653
• SNOW-3243341: Python decorators for required conditions in cursor #696
• SNOW-3243341: optional dependencies in cursor (pyarrow, pandas) #639
• SNOW-3243341: e2e tests for pyarrow & pandas type conversion #629
• SNOW-3243341: Python e2e tests for pyarrow and pandas support #625
• main

This stack of pull requests is managed by Graphite. Learn more about stacking.

[sfc-gh-turbaszek]

Does it pass with reference driver?

[sfc-gh-asolarski]

fixed (it seems that None incorrect per pep249 but backward compatible)
I covered this additionally with more integration tests.

[sfc-gh-turbaszek]

Suggested change

	def create_statement(connection: Connection, query: str) -> Generator[StatementHandle]:
	Query = str
	def create_statement(connection: Connection, query: Query) -> Generator[StatementHandle]:

Nit, we can introduce Query = str it would be aligned with Row = tuple[Any, ...] DictRow = dict[str, Any]. Possibly having snowflake.connector.types can be reasonable place to keep all custom types. Not mandatory in this PR

[sfc-gh-asolarski]

It looks like a good idea indeed.
I will take a look into that in one of the following PRs.