feat: support different USE statement syntaxes

[kacpermuda]

I'd like to add support for different syntaxes of the USE statement, used to set a default DATABASE/SCHEMA for set of queries.

Some implementations are calling their top level structure CATALOG (e.g. Databricks, Trino), but most of them stick to DATABASE so I assumed that we should stick to naming: top-level container = DATABASE, lower-level container = SCHEMA. For the full identifier we end up with: DATABASE.SCHEMA.TABLE.

In some implementations DATABASE and SCHEMA are aliases, but usually these implementations have a two-layer structure (DATABASE.TABLE, without SCHEMA) - in that case we should keep DATABASE as DATABASE and leave SCHEMA empty.

There are different syntaxes of USE statement across dialects, but I implemented it like:
(use syntax => resolved db and schema names)

1. USE x; -> DATABASE=x; SCHEMA=null; (Databricks is adjusted to DATABASE=null; SCHEMA=x;)
2. USE x.y; -> DATABASE=x; SCHEMA=y;
3. USE DATABASE x; -> DATABASE=x; SCHEMA=null; (Databricks is adjusted to DATABASE=null; SCHEMA=x; In Databricks DATABASE is an alias for SCHEMA and they both mean lower level structure while the top level structure is called CATALOG)
4. USE SCHEMA y; -> DATABASE=null; SCHEMA=y;
5. USE SCHEMA x.y; -> DATABASE=x; SCHEMA=y; (Snowflake specific syntax, SCHEMA keyword is optional and should not influence the parsing of fully qualified name)
6. USE CATALOG x; -> DATABASE=x; SCHEMA=null; (Databricks-specific keyword: CATALOG)
7. USE DEFAULT; -> DATABASE=null; SCHEMA=null; (Hive specific, means going back to default database set - we don't know what it is just by looking at the query.)

Docs about USE statement in different dialects:

• MySQL 8.4 USE
• MySQL 5.7 USE
• Snowflake USE SCHEMA
• Snowflake USE DATABASE
• SQL Server USE
• ClickHouse USE
• DuckDB USE
• Hive USE
• Databricks USE CATALOG
• Databricks USE SCHEMA
• Databricks USE DATABASE

Some comments:

• Bigquery, PostgreSQL, Redshift, SQLite do not allow USE statements;

[iffyio]

LGTM! cc @alamb

[alamb]

Thank you @kacpermuda and @iffyio

This change makes sense to me

[coveralls]

Pull Request Test Coverage Report for Build 10509933459

Warning: This coverage report may be inaccurate.

This pull request's base commit is no longer the HEAD commit of its target branch. This means it includes changes from outside the original pull request, including, potentially, unrelated coverage changes.

• For more information on this, see Tracking coverage changes with pull request builds.
• To avoid this issue with future PRs, see these Recommended CI Configurations.
• For a quick fix, rebase this PR at GitHub. Your next report should be accurate.

Details

• 186 of 189 (98.41%) changed or added relevant lines in 10 files are covered.
• No unchanged relevant lines lost coverage.
• Overall coverage increased (+0.05%) to 89.191%

---

Changes Missing Coverage	Covered Lines	Changed/Added Lines	%
src/ast/dcl.rs	8	9	88.89%
src/parser/mod.rs	15	16	93.75%
tests/sqlparser_hive.rs	14	15	93.33%

Totals
Change from base Build 10418247467:	0.05%
Covered Lines:	28468
Relevant Lines:	31918

---

💛 - Coveralls

[alamb]

🚀