[Query] Implement GROUP BY, Aggregates and HAVING clauses

[remiceres]

Summary

Implement SPARQL 1.1 solution sequence partitioning (GROUP BY), aggregate computation (COUNT, SUM, AVG, MIN, MAX, SAMPLE, GROUP_CONCAT), and post-aggregation filtering (HAVING) in core.next.query.

---

Objectives and Technical Scope

1. Solution Sequence Grouping (GROUP BY)

• Partition incoming solution sequences into multiset groups according to key variables or expressions.
• Support implicit single-group aggregation when aggregate functions are present without an explicit GROUP BY.

2. Aggregate Functions

• Implement stateful aggregate evaluators:
  • COUNT(?var), COUNT(*), COUNT(DISTINCT ?var).
  • SUM(?var), SUM(DISTINCT ?var).
  • AVG(?var), AVG(DISTINCT ?var).
  • MIN(?var), MAX(?var).
  • SAMPLE(?var).
  • GROUP_CONCAT(?var; separator="..."), GROUP_CONCAT(DISTINCT ?var; separator="...").

3. Post-Aggregation Filtering (HAVING) and Projection Rules

• Evaluate boolean filter expressions over computed group aggregate states in HAVING.
• Enforce SPARQL 1.1 projection constraints (only grouped variables or aggregate expressions in SELECT).

4. Validation and Conformance

• Validate against official W3C SPARQL 1.1 test suites (aggregates, grouping, having) in corese-w3c.

---

Dependencies

• Blocked by:
  • [Query] Implement SPARQL 1.1 scalar built-in functions and casting operators #563
  • [Query] Implement Subqueries, EXISTS, NOT EXISTS and MINUS operators #564
• Blocks:
  • [Update] Implement SPARQL 1.1 Update execution engine (INSERT, DELETE, LOAD, CLEAR) #567