fix(mysql): preserve DECIMAL precision when scale is null

[Aias00]

Related issue

Closes #2180

Summary

MysqlColumnTypeEnum.buildDataType guarded with if (columnSize == null || decimalDigits == null), returning bare DECIMAL for DECIMAL(10) (precision set, scale null) — the precision was silently lost (MySQL defaults to DECIMAL(10,0)). The size-only branch was unreachable dead code. Changed the guard to columnSize == null only, making the size-only branch reachable. Mirrors every sibling (Oracle/DM/DB2/Oscar). Same fix as the already-merged Sqlite #2167.

Verification

• mvn compile -> BUILD SUCCESS.

Contributor declaration

• I linked the Issue that defines this change.
• I tested the affected behavior and reported the actual results above.
• I did not include credentials, private data, or generated build output.
• I disclosed substantial AI assistance below, or this PR contains no substantial AI-generated code.

AI assistance: The fix, verification, and PR description were produced with Claude Code assistance.

[Copilot]

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.

[Aias00]

Live-database verification (MySQL 8)

• DECIMAL(15) → MySQL shows decimal(15,0) ✅ (precision 15 preserved)
• DECIMAL(15,2) → decimal(15,2) ✅
• Bare DECIMAL → decimal(10,0) (MySQL default)

Before the fix: DECIMAL(15) with null scale → dead branch returned bare DECIMAL → MySQL defaulted to decimal(10,0) — user's precision 15 was silently downgraded to 10.

[openai0229]

Thanks for the update. This is not ready to merge in its current form.

The DECIMAL(P) direction is reasonable, but this change applies to both broad branches in buildDataType, so it also changes generated SQL for FLOAT, DOUBLE, and UNSIGNED variants. The current verification only covers DECIMAL, and does not establish that one-argument precision is valid for every affected MySQL type.

Please either narrow the fix to the intended types or add type-by-type automated generated-DDL tests against MySQL, including the one-argument DOUBLE behavior and signed/unsigned cases. Please also rebase onto the latest main before requesting another review.