issue #7488 : Sorted Schema Merge transform

[mattcasters]

What changes were proposed in this pull request?

Add a Sorted Schema Merge transform (issue #7488), ported from the hop-data-vault effort.

It combines Sorted Merge and Stream Schema Merge:

• Merges multiple pre-sorted input streams into one globally sorted output
• Supports different schemas per input and produces a unified sparse row layout
• Useful with Repeat Fields when combining multi-source / Data Vault satellite data into SCD2-style streams

Also includes:

• Unit tests for schema union, sort-key resolution, and compare logic
• User manual docs and transform icon
• Integration test main-0106-sorted-schema-merge with a golden sparse-row data set

What was tested?

• ./mvnw -pl plugins/transforms/sortedschemamerge spotless:apply test (5 unit tests)
• Ran main-0106-sorted-schema-merge.hwf via hop-run against a local Hop client with the new plugin installed; unit test and golden data set validation passed

How can this be tested?

1. Build and install the sortedschemamerge transform plugin
2. Create a pipeline with 2+ pre-sorted streams of different layouts → Sorted Schema Merge (sort key) → Dummy
3. Or run the transforms integration test:
   • main-0106-sorted-schema-merge.hwf / unit test 0106-sorted-schema-merge UNIT

Fixes: #7488