Align TableMode and LogLevel enums between frontend and backend

[Copilot]

• Analyze existing state of frontend and backend enums
• Add SKIP to backend TableMode enum (TableConfiguration.kt)
• Add DEBUG to backend LogLevel enum (JobLog.kt)
• Handle TableMode.SKIP in JobService.kt — short-circuit the table loop to skip both schema mirroring and data processing
• Expand EnumAlignmentTest.kt to cover all enum alignments
• Run backend tests — all 7 EnumAlignmentTest tests pass

[Copilot]

Pull request overview

Aligns backend enum definitions with the frontend to prevent enum deserialization/persistence failures (HTTP 400s) when the UI sends newer enum values, and strengthens CI coverage to catch future enum drift.

Changes:

• Added missing enum values on backend: TableMode.SKIP and LogLevel.DEBUG.
• Updated job processing logic to handle TableMode.SKIP explicitly.
• Strengthened EnumAlignmentTest to assert exact canonical enum value sets across multiple enums.

Reviewed changes

Copilot reviewed 4 out of 4 changed files in this pull request and generated 1 comment.

File	Description
backend/src/main/kotlin/com/opendatamask/model/TableConfiguration.kt	Adds SKIP to TableMode to match frontend values.
backend/src/main/kotlin/com/opendatamask/model/JobLog.kt	Adds DEBUG to LogLevel to match frontend values.
backend/src/main/kotlin/com/opendatamask/service/JobService.kt	Adds SKIP handling branch when processing a table.
backend/src/test/kotlin/com/opendatamask/model/EnumAlignmentTest.kt	Replaces a weak containment assertion with exact-set assertions for several enums.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

TableMode.SKIP only skips the data-processing step in processTable, but runJob still calls destinationSchemaService.mirrorSchema(...) for every table before processTable runs. As a result, a “skipped” table will still have its schema created/updated in the destination, which contradicts the log message and likely the intent of SKIP. Consider short-circuiting earlier (e.g., in the table loop) so SKIP avoids schema mirroring as well, or adjust the logging/semantics to make it explicit that schema is still mirrored.

Suggested change

	addLog(jobId, "Skipping table: ${tableConfig.tableName}", LogLevel.INFO)
	addLog(
	jobId,
	"Skipping data processing for table: ${tableConfig.tableName} (schema may still be mirrored)",
	LogLevel.INFO
	)

[MaximumTrainer]

@copilot apply changes based on the comments in this thread

[Copilot]

@copilot apply changes based on the comments in this thread

Done in dbb638a. The SKIP check now short-circuits at the top of the table loop in runJob, so both mirrorSchema and processTable are bypassed for skipped tables. The SKIP branch in processTable is kept (Kotlin requires it for exhaustiveness) but marked unreachable.