🚀 elevata 1.1.0 — Query Builder Governance & UNION Workflow

This release is a major evolution of the Query Builder and metadata engine.

It introduces a full, metadata-driven UNION workflow and significantly improves
governance, safety, and determinism across the platform.

---

✨ What’s new

🛠️ Query Builder: UNION done right

You can now build and validate UNIONs with full transparency:

• Define multiple UNION branches
• Generate and manage a shared output schema
• Map each branch to the output contract
• Auto-map by name
• Validate and set as final query head

All of this is available through a guided, scoped UI – no SQL required.

---

🛡️ Stronger governance

• Clear separation of query_root (anchor) and query_head (final output)
• Head defines:
  • SQL preview
  • Output schema contract
  • Validation scope
• Guardrails prevent destructive changes when downstream datasets depend on a contract.

---

🪶 Safer historization

• Renames are tracked via former_names and propagated correctly.
• Dropped rawcore columns are preserved in _hist as inactive orphans.
• FK reference deletion can never remove hist columns.
• Technical SCD columns always stay at the end of the table.

---

🧠 Determinism by design

• ORDER BY / PARTITION BY governance for aggregates and windows
• Deterministic execution checks in the Query Builder
• Contract diffs highlight breaking changes early

---

💡 Why this matters

elevata is no longer just a SQL generator.
It is now a metadata-driven, deterministic query engine
with first-class contracts, lineage, and governance.

This release lays the foundation for safe, explainable analytics at scale.

Full Changelog: v1.0.0...v1.1.0