fix: make query_ids in SnowflakeSqlApiOperator in deferrable mode consistent

[kacpermuda]

Few fixes included in this PR:

1. SnowflakeSqlApiHook.execute_query will now consistently reset query_ids each time it's called, exactly like SnowflakeHook.run does. Currently the behavior depended on number of queries executed, which could cause getting query_ids from previous execution as an attribute - sometimes it was overwritten, sometimes not.
2. SnowflakeSqlApiTrigger will now return the query_ids of query actually executed, instead of child query_ids of multi-statement queries, which is consistent with non-deferrable behavior that we have right now. This change makes the behavior consistent regardless of execution mode.
3. I've added re-assignment of query_ids after coming back from deferral in SnowflakeSqlApiOperator, so that the listeners can use that information and have consistent behavior regardless of execution mode.
4. I've added an explicit caching of SnowflakeSqlApiHook in SnowflakeSqlApiOperator, right now the property of super class SQLExecuteQueryOperator was overwritten during execution, which could lead to inconsistencies, depending on when you ask for SnowflakeSqlApiOperator._hook.
5. Added missing self.query_ids definition in init of SnowflakeSqlApiOperator

Added tests for all changes.

---

^ Add meaningful description above
Read the Pull Request Guidelines for more information.
In case of fundamental code changes, an Airflow Improvement Proposal (AIP) is needed.
In case of a new dependency, check compliance with the ASF 3rd Party License Policy.
In case of backwards incompatible changes please leave a note in a newsfragment file, named {pr_number}.significant.rst or {issue_number}.significant.rst, in airflow-core/newsfragments.