[SPARK-52159][SQL] Properly handle table existence check for jdbc dialects

[milastdbx]

What changes were proposed in this pull request?

In this PR, I propose that we rethrow exception when we are doing table existence check in jdbc dialect, if exception is not related to table/schema not being found. I propose this because currently all exceptions get swallowed and method returns false. From the perspective of the system its as if table doesn't exist, which is a wrong message (e.g, we can get table does not exist even if it was network failure).

This issue is mostly exposed when TableCatalog API is used,

  override def loadTable(ident: Identifier): Table = {
    if (!tableExists(ident)) {
      throw QueryCompilationErrors.noSuchTableError(ident)
    }

Error code links:

• https://www.ibm.com/docs/en/db2-for-zos/12.0.0?topic=codes-error-sql
• https://db.apache.org/derby/docs/10.4/ref/rrefexcept71493.html
• https://docs.databricks.com/aws/en/error-messages/sqlstates
• https://www.h2database.com/javadoc/org/h2/api/ErrorCode.html
• https://learn.microsoft.com/en-us/sql/relational-databases/errors-events/database-engine-events-and-errors-0-to-999?view=sql-server-ver16
• https://dev.mysql.com/doc/mysql-errors/8.0/en/server-error-reference.html
• https://www.postgresql.org/docs/current/errcodes-appendix.html

How its implemented today, tableExist cannot throw anything, so every exception gets converted to noSuchTableError which is wrong.

Does this PR introduce any user-facing change?

Customers will get proper error messages.

How was this patch tested?

Tests

Was this patch authored or co-authored using generative AI tooling?

Generetad by: COPILOT

[milastdbx]

I will create SPARK jira ticket shortly

[MaxGekk]

@milastdbx Could you fix the test failure, please. I think it is related to your changes:

[info] JDBCTableCatalogSuite:
[info] - show tables (61 milliseconds)
[info] - drop a table and test whether the table exists *** FAILED *** (86 milliseconds)
[info]   "[SCHEMA]_NOT_FOUND" did not equal "[TABLE_OR_VIEW]_NOT_FOUND" (SparkFunSuite.scala:346)
[info]   Analysis:
[info]   "[SCHEMA]_NOT_FOUND" -> "[TABLE_OR_VIEW]_NOT_FOUND"

Also add the [SQL] tag to PR's title.

[alekjarmov]

Would be nice to add comments with the relevant error code links in the override function like here

[MaxGekk]

Would be nice to add comments with the relevant error code links

It would be nice to have the links but they can become outdated, unfortunately, and this is out of our control.

+1, LGTM. Merging to master.
Thank you, @milastdbx and @alekjarmov for review.