UW3 release candidate

.claude/commands/check-patterns.md

@@ -0,0 +1,69 @@
+---
+description: Find deprecated access patterns in code and documentation
+---
+
+## Reference Document
+
+**Read the authoritative patterns guide first:**
+`docs/developer/UW3_Style_and_Patterns_Guide.qmd`
+
+Key sections:
+- Section 4 "Array and Data Management" - array indexing patterns
+- Section 5 "Context Managers" - deprecated vs current access patterns
+- Section 6 "MPI and Parallel Patterns" - parallel safety
+
+---
+
+## Search for Legacy Patterns
+
+### Source Code
+```bash
+grep -r "with.*\.access(" src/underworld3/ --include="*.py"
+```
+
+### Examples
+```bash
+grep -r "with.*\.access(" docs/examples/ --include="*.py"
+grep -r "mesh\.data\[" docs/examples/ --include="*.py"
+```
+
+### Documentation (markdown, quarto)
+```bash
+grep -r "with.*\.access(" docs/ --include="*.md" --include="*.qmd"
+grep -r "mesh\.data\[" docs/ --include="*.md" --include="*.qmd"
+```
+
+### Notebooks
+```bash
+find docs -name "*.ipynb" -exec grep -l "\.access(" {} \;
+find docs -name "*.ipynb" -exec grep -l "mesh\.data" {} \;
+```
+
+---
+
+## Pattern Classification
+
+| Pattern | Status | Replacement |
+|---------|--------|-------------|
+| `with mesh.access(var):` | **Deprecated** | Direct: `var.data[...]` |
+| `with swarm.access(var):` | **Deprecated** | Direct: `var.data[...]` |
+| `mesh.data` (coordinates) | **Deprecated** | `mesh.X.coords` |
+| `var.array[:, 0, 0]` | Current | Scalar indexing |
+| `var.array[:, 0, :]` | Current | Vector indexing |
+| `uw.synchronised_array_update()` | Current | Multi-variable batch |
+
+---
+
+## Analysis Required
+
+For each occurrence found, classify as:
+- **Safe to remove**: Simple data access, visualization, I/O
+- **Preserve for now**: Solver-critical operations needing further analysis
+- **Already correct**: Part of the access context implementation itself
+
+## Output Format
+
+- Total occurrences by location (src, examples, docs, notebooks)
+- Files needing updates grouped by type
+- Priority order (user-facing examples first)
+- Recommended fixes

.claude/commands/test-regression.md

@@ -0,0 +1,24 @@
+---
+description: Check regression test suite status
+---
+
+Run the regression test suite and report on any failures:
+
+```bash
+pixi run -e default pytest tests/test_06*_regression.py -v --tb=short
+```
+
+## Analysis Required
+
+- Count total tests and number passing/failing
+- For each failure, identify:
+  - Test name and location
+  - Error type (AttributeError, AssertionError, etc.)
+  - Root cause category (API assumption issue, actual bug, naming convention)
+- Provide summary with recommended next actions
+
+## Output Format
+
+- Summary: X/Y tests passing
+- Failures grouped by category
+- Recommended fixes for each failure type

.claude/commands/test-solvers.md

@@ -0,0 +1,39 @@
+---
+description: Validate complex solver tests
+---
+
+Run the complex solver test suite to validate core physics functionality:
+
+```bash
+# Test Stokes solvers
+pixi run -e default pytest tests/test_101*_Stokes*.py -v --tb=short
+
+# Test advection-diffusion solvers
+pixi run -e default pytest tests/test_110*_AdvDiff*.py -v --tb=short
+
+# Test other complex systems
+pixi run -e default pytest tests/test_11*_*.py -v --tb=short
+```
+
+## Analysis Required
+
+- Report pass/fail counts for each category (Stokes, AdvDiff, Other)
+- For failures, categorize by:
+  - Solver type
+  - Error pattern (convergence, numerical accuracy, API usage)
+- Identify if failures are related to recent changes
+
+## Critical Note
+
+These tests validate core physics. Any failures are **high priority**.
+
+## Output Format
+
+| Test Category | Total | Passed | Failed |
+|---------------|-------|--------|--------|
+| Stokes        |       |        |        |
+| AdvDiff       |       |        |        |
+| Other         |       |        |        |
+
+- Detailed failure analysis grouped by solver type
+- Assessment: Are failures new regressions or pre-existing?

.claude/commands/test-tier-a.md

@@ -0,0 +1,35 @@
+---
+description: Run only Tier A (production-ready) tests - trusted for TDD and CI
+---
+
+Run only the most reliable, production-ready tests:
+
+```bash
+pixi run -e default pytest -m "tier_a" -v --tb=short
+```
+
+## What This Tests
+
+**Tier A tests are:**
+- Long-lived tests with proven track record (>3 months)
+- Consistently passing across multiple environments
+- Testing stable, well-understood functionality
+- Failure indicates **DEFINITE regression** in production code
+
+## When to Use
+
+- Test-Driven Development (TDD) sprints
+- Continuous Integration (CI) pipelines
+- Release validation
+- Bisecting regressions (these tests can be trusted)
+
+## If a Tier A Test Fails
+
+**HIGH PRIORITY** - This indicates a real regression:
+1. Investigate immediately
+2. Bisect to find breaking commit
+3. Fix production code (or demote test if incorrectly promoted)
+
+## Reference
+
+See `docs/developer/TESTING-RELIABILITY-SYSTEM.md` for tier definitions.

.claude/commands/test-tier-ab.md

@@ -0,0 +1,31 @@
+---
+description: Run Tier A+B tests - full validation suite
+---
+
+Run production-ready and validated tests (excludes experimental):
+
+```bash
+pixi run -e default pytest -m "tier_a or tier_b" -v --tb=short
+```
+
+## What This Tests
+
+**Tier A:** Production-ready (failures = definite regression)
+**Tier B:** Validated but newer (failures need investigation)
+
+## When to Use
+
+- Full validation before merging
+- Feature completion verification
+- Manual review of new functionality
+
+## Interpreting Results
+
+| Tier | If Fails | Action |
+|------|----------|--------|
+| A    | HIGH priority | Investigate immediately - real regression |
+| B    | MEDIUM priority | Could be test OR code issue - investigate |
+
+## Reference
+
+See `docs/developer/TESTING-RELIABILITY-SYSTEM.md` for tier definitions.

.claude/commands/test-units-classify.md

@@ -0,0 +1,45 @@
+---
+description: Analyze units test failures and classify for reliability tiers
+---
+
+Run units tests and classify any failures by reliability tier:
+
+```bash
+pixi run -e default pytest tests/test_07*_units*.py tests/test_08*_*.py -v --tb=short 2>&1 | head -200
+```
+
+## Classification Criteria
+
+### Tier A Candidates (Production-Ready)
+- [ ] Test has existed for >3 months
+- [ ] Test passes consistently
+- [ ] Functionality is stable and production-used
+- [ ] Failure would indicate definite bug
+
+### Tier B Candidates (Validated)
+- [ ] Test passes at least once
+- [ ] Functionality appears correct
+- [ ] Needs more production validation
+- [ ] Test is new (<3 months)
+
+### Tier C Candidates (Experimental)
+- [ ] Feature is not fully implemented
+- [ ] Test documents expected future behavior
+- [ ] Known issues exist
+
+## Analysis Required
+
+For each failing test:
+1. Identify root cause (test issue vs code issue)
+2. Classify into appropriate tier
+3. Recommend: fix, mark xfail, or remove
+
+## Output Format
+
+| Test File | Status | Tier | Recommendation |
+|-----------|--------|------|----------------|
+| test_0700_* | Pass/Fail | A/B/C | Action needed |
+
+## Reference
+
+See `docs/developer/TEST-CLASSIFICATION-2025-11-15.md` for current status.

.claude/commands/test-units.md

@@ -0,0 +1,29 @@
+---
+description: Quick validation of units system
+---
+
+Run the complete units test suite to ensure dimensional analysis is working:
+
+```bash
+pixi run -e default pytest tests/test_07*_units*.py tests/test_08*_*.py -v --tb=short
+```
+
+## Expected Results
+
+Most tests should pass. Check `docs/developer/TEST-CLASSIFICATION-2025-11-15.md` for current baseline.
+
+## Analysis Required
+
+If failures occur, identify which subsystem:
+- **Core units** (test_0700): Fundamental units system
+- **Unit conversion** (test_0801): Utility functions
+- **Workflow integration** (test_0803): End-to-end workflows
+- **Stokes ND** (test_0818): Non-dimensional solver integration
+- **Coordinate units**: Dimensional analysis of mesh coordinates
+
+## Output Format
+
+- Quick status: X/Y tests passing
+- If failures: detailed breakdown by subsystem
+- Assessment: regression vs known issue
+- Recommended fixes if needed