Initial separation of Snowflake_Dialect from Postgres_Dialect

[radeusgd]

Pull Request Description

• Part of #9486
• Building on top of initial work by @jdunkerley and finishing it
  • Reverted the changes to the Postgres_Dialect from last Snowflake work and split the Snowflake_Dialect into a separate module.
  • Moved from rounding_decimal_places_not_allowed_for_floats to supports_float_round_decimal_places (as too confusing).
  • Added Snowflake_Dialect type.
• Extracted Snowflake_Spec into separate Snowflake_Tests
  • It imports the common tests from Table_Tests.
• Some initial adaptations to make the snowflake dialect not-crash.
  • Adding Internals_Access proxy to allow external implementations to access our internal data structures without directly exposing them to users. Users should not use these.
  • Adding profiling of SQL to check performance.

Important Notes

Checklist

Please ensure that the following checklist has been satisfied before submitting the PR:

• [x] The documentation has been updated, if necessary.
• [ ] Screenshots/screencasts have been attached, if there are any visual changes. For interactive or animated visual changes, a screencast is preferred.
• [x] All code follows the Scala, Java, TypeScript, and Rust style guides. In case you are using a language not listed above, follow the Rust style guide.
• [x] Unit tests have been written where possible.

[radeusgd]

Trying to run the Snowflake tests. At first it was failing miserably and completely crashing mid-way.

After some fixes I'm able to run the suite, although not everything works. Still I think it's enough to integrate - we can work on fixing tests later.

We won't be running these on CI yet - they are far too slow and need a separate workflow, probably not run for most PRs. So it's ok to keep them failing - for many tests it's similar amount of work to mark them pending as to fix them.

Current status:

• the test suite runs to the end.
• 927 tests succeed and 493 fail. Some are marked pending="TODO" to be revised.
• the majority of failures stems from the fact that Snowflake does not return the results in the 'insertion-order'. This is valid SQL behaviour - without ORDER BY the ordering is unspecified. However, since our current backends actually had deterministic ordering, lots of our tests relied on this implicit assumption. We will need to adapt most of these to avoid this assumption as it is simply incorrect.
• other big group of tests is failures related to date/time types - I did not yet revise the type mapping - this will be done in follow up.
• there are some tests that fail due to unknown methods - still using method names from Postgres in some places. To be addressed later.

The biggest issue is the execution time of tests. Running the whole suite takes 45 minutes (!!!). In comparison running SQLite or Postgres tests takes ~2.5 minutes each.

I tried writing some profiling code and it turns out the problem is that we are running lots of queries (as we are testing lots of stuff). The Snowflake query latency is actually very nice - averaging at 187 ms per operation - for a remote Database operation that's not too bad. However, in local instances of Postgres or SQLite it's just around 1 ms - so the difference is huge - roughly 100x. Given that even the lightening-fast local tests run for about 2 minutes, it is not a surprise that the 100x slower remote tests take up to an hour... This may be a problem for other remote backends as well.

[radeusgd]