Experiment: atomic target-first chained search with sort support

[Mikael Weaver (mikaelweave)]

Comparison goal

This draft compares a cleaner atomic target-first chained-search rewrite against PR #5725's narrowly gated implementation.

Changes

• recognizes and emits the complete target-first CTE group at the root query-generator level
• owns selective target reference, source backlink, source date predicate, and an immediately following sort projection as one atomic rewrite
• preserves ascending/descending plain Sort and SortWithFilter behavior, Min/Max selection for multi-valued rows, deduplication, and continuation-token tie breaks
• preserves missing-value two-phase sort selection, Top, count-only, include/revinclude, and trailing-filter propagation
• uses an explicit match result with consumed-expression count and returned CTE IDs instead of out parameters, magic return counts, and counter-derived IDs
• extracts repeated status-filter test setup

Intentional limitation

A sort separated from the target-first predicates by another source filter is deliberately not consumed atomically. The optimized three-CTE filter feeds the existing generic filter/sort pipeline so the intervening predicate is applied before sort projection; this avoids partial semantic changes while retaining the target-first optimization.

Validation

• baseline SQL Server unit test project build passed before changes
• updated SQL Server unit test project build passed with 0 warnings and 0 errors
• focused SqlQueryGeneratorTests passed: 35/35
• isolated CompartmentQueryGeneratorTests passed: 26/26 after an unrelated transient full-suite failure
• full Microsoft.Health.Fhir.SqlServer.UnitTests suite passed: 1,072/1,072
• rubber-duck review found no correctness or simplicity issues

[Codecov Comments Bot (codecov-commenter)]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 77.38%. Comparing base (b3f01fa) to head (76dbbd4).

Additional details and impacted files

@@                          Coverage Diff                           @@
##           fix/selective-first-chained-search    #5726      +/-   ##
======================================================================
+ Coverage                               77.25%   77.38%   +0.13%     
======================================================================
  Files                                    1012     1012              
  Lines                                   37393    37415      +22     
  Branches                                 5690     5689       -1     
======================================================================
+ Hits                                    28887    28953      +66     
+ Misses                                   7131     7098      -33     
+ Partials                                 1375     1364      -11     

see 15 files with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.