fix(scan): use warehouse-backed direct scans

[jonathanhaaswriter]

Summary

• switch direct scan and scan preflight from Snowflake-only wiring to the configured warehouse so sqlite/postgres backends can run scans without a Snowflake client
• add shared warehouse asset-query helpers that preserve _cq_table, dedupe on latest _cq_sync_time, and honor incremental/keyset scan cursors on sqlite/postgres
• add regression coverage for warehouse-backed direct scans plus sqlite incremental pagination/filter behavior

Validation

• go test ./...
• GOTOOLCHAIN=go1.26.1 go run github.com/golangci/golangci-lint/v2/cmd/golangci-lint@v2.8.0 run --timeout 5m ./...
• go run ./scripts/check_agent_sdk_contract_compat/main.go --require-baseline --base-ref=writer/main
• make agent-sdk-docs-check
• make agent-sdk-packages-check
• go run ./scripts/check_api_contract_compat/main.go --require-baseline --base-ref=writer/main
• make api-contract-docs-check
• make config-docs-check
• make devex-codegen-check
• go test ./internal/app -run "Test(PreCommitHookRunsFastLintOnStagedGoFiles|PrePushHookRunsChangedDevexPreflight|DockerBuildCommandsPassGoVersionBuildArg|DevelopmentGuideDocumentsDevexPreflight|DevexScriptPlansRelevantChecks|DevexScriptChangedModeIncludesWorkspaceDiffSources)"
• make openapi-check
• go run ./scripts/check_report_contract_compat/main.go --require-baseline --base-ref=writer/main
• make report-contract-docs-check
• make vendor-check

[jonathanhaaswriter]

I think the new direct-scan path loses watermark progress for delete-only incremental batches. The current flow exits early when GetAssets returns no rows and only persists watermarks when scanned > 0, but delete-only CDC batches still need to advance the cursor even when there is nothing left to fetch. Otherwise the scan can keep replaying the same delete-only window.

Can we preserve the old "advance watermark from change events even when no assets are returned" behavior?