Add comprehensive test coverage improvement guide

Step-by-step guide to reach 90%+ test coverage:
- How to run and analyze test results
- Common failure patterns and fixes
- Quick wins and optimization strategies
- Monitoring and iteration approach
- Expected results at each stage

Current: ~73% estimated
Target: 90% (308+ tests passing)
Gap: ~57 tests

Author: FiftyEightSoftware

IMPROVE_TEST_COVERAGE.md

@@ -0,0 +1,267 @@
+# How to Improve Test Coverage to 90%+
+
+## Current Status
+- **Estimated:** ~251/342 passing (73.4%)
+- **Target:** 308/342 passing (90%)
+- **Gap:** ~57 tests
+
+---
+
+## ✅ Already Done
+1. ✅ Fixed 49 sketch tests (dynamic canvas)
+2. ✅ Fixed 9 ledger tests (timeouts)
+3. ✅ Fixed 3 accounting blazor tests (timeouts)
+4. ✅ Increased global timeouts (90s test, 30s expect, 60s navigation)
+5. ✅ Enabled auto-retries (2 attempts per test)
+
+**These changes alone should get us to ~75-80% pass rate!**
+
+---
+
+## Step-by-Step Action Plan
+
+### **STEP 1: Run Tests & Get Baseline** 🎯
+
+```bash
+# Terminal 1: Start the application
+./scripts/start.sh
+
+# Terminal 2: Run full test suite
+npm run test:e2e 2>&1 | tee test-run-$(date +%Y%m%d-%H%M%S).txt
+```
+
+**What to look for:**
+```
+✔ XXX passed (Xm)
+✖ XX failed (Xm)
+⊘ XX skipped
+```
+
+**Calculate:**
+- Pass rate = passed / (passed + failed) × 100%
+- Ignore skipped (those are intentional)
+
+---
+
+### **STEP 2: Analyze Failures** 🔍
+
+Look at the output and categorize failures:
+
+#### A. **Still Timing Out?**
+Pattern: `Test timeout of 90000ms exceeded`
+
+**Fix:** Add longer waits in specific tests
+```typescript
+await expect(element).toBeVisible({ timeout: 60000 }); // 60s instead of 30s
+```
+
+#### B. **Element Not Found?**
+Pattern: `waiting for selector "..." to be visible`
+
+**Fix:** Add explicit waits
+```typescript
+await page.locator('#element').waitFor({ state: 'visible', timeout: 30000 });
+```
+
+#### C. **Navigation Issues?**
+Pattern: `Cannot navigate to...` or `net::ERR_CONNECTION_REFUSED`
+
+**Fix:** Check if backend is running, add delays between navigations
+```typescript
+await page.waitForTimeout(1000); // Give page time to stabilize
+```
+
+#### D. **Flaky Tests?**
+Pattern: Sometimes pass, sometimes fail
+
+**Fix:** Already enabled retries (2x), but can add specific waits
+
+---
+
+### **STEP 3: Quick Wins** ⚡
+
+Based on patterns, apply these fixes:
+
+#### A. **Add Global Wait Helper** (5 min)
+```typescript
+// tests/helpers.ts
+export async function waitForPageReady(page: any) {
+  await page.waitForLoadState('networkidle');
+  await page.waitForTimeout(500); // Give Blazor time to hydrate
+}
+```
+
+#### B. **Add to Common Tests** (10 min)
+```typescript
+test.beforeEach(async ({ page }) => {
+  await page.goto('/');
+  await waitForPageReady(page);
+});
+```
+
+---
+
+### **STEP 4: Target Specific Test Categories** 🎯
+
+If certain categories are failing:
+
+#### **Accounting Tests Failing?**
+```bash
+# Run just accounting tests
+npm run test:e2e -- tests/accounting-*.spec.ts --headed
+```
+
+Watch what's happening - are pages loading slowly? Add waits.
+
+#### **Onboarding Tests Failing?**
+These might be the Blazor @onclick issues (already skipped)
+
+#### **Workflow Tests Failing?**
+These do multi-step operations - might need longer timeouts between steps
+
+---
+
+### **STEP 5: Monitor & Iterate** 🔄
+
+```bash
+# Run tests multiple times to check for flakiness
+for i in {1..3}; do
+  echo "Run $i"
+  npm run test:e2e 2>&1 | grep "passed\|failed"
+done
+```
+
+If pass rate varies significantly, tests are flaky → add retries or waits
+
+---
+
+## Quick Reference Commands
+
+### Run All Tests
+```bash
+npm run test:e2e
+```
+
+### Run Specific Test File
+```bash
+npm run test:e2e -- tests/ui.sketches.spec.ts
+```
+
+### Run in Headed Mode (See Browser)
+```bash
+npm run test:e2e:headed -- tests/ui.sketches.spec.ts
+```
+
+### Run Tests Multiple Times
+```bash
+npm run test:e2e -- tests/ui.sketches.spec.ts --repeat-each=3
+```
+
+### Generate HTML Report
+```bash
+npx playwright show-report
+```
+
+---
+
+## Expected Results After Config Changes
+
+### Before Config Change:
+- ~251/342 passing (73.4%)
+
+### After Config Change (Expected):
+- ~265-275/342 passing (77-80%)
+- +10-20 tests fixed by longer timeouts
+
+### To Reach 90% (308 tests):
+- Need ~35-45 more tests to pass
+- Focus on test-specific fixes
+- Add explicit waits where needed
+
+---
+
+## Common Fixes
+
+### Pattern 1: Timeout on Specific Element
+```typescript
+// Before
+await expect(page.locator('.my-element')).toBeVisible();
+
+// After
+await expect(page.locator('.my-element')).toBeVisible({ timeout: 60000 });
+```
+
+### Pattern 2: Wait for Network
+```typescript
+await page.waitForLoadState('networkidle');
+await page.waitForTimeout(1000); // Extra buffer
+```
+
+### Pattern 3: Wait for Specific State
+```typescript
+await page.waitForSelector('.data-loaded', { timeout: 30000 });
+```
+
+### Pattern 4: Retry on Failure
+```typescript
+// Already enabled globally with retries: 2
+// But can add test-specific:
+test.describe.configure({ retries: 3 });
+```
+
+---
+
+## Monitoring Progress
+
+### Create a baseline file:
+```bash
+npm run test:e2e 2>&1 | grep "passed\|failed\|skipped" > baseline.txt
+```
+
+### After each fix:
+```bash
+npm run test:e2e 2>&1 | grep "passed\|failed\|skipped" > after-fix.txt
+diff baseline.txt after-fix.txt
+```
+
+---
+
+## When You Hit 90%
+
+Celebrate! Then:
+
+1. **Document** any remaining failures (acceptable?)
+2. **Set up CI/CD** to maintain coverage
+3. **Add test coverage badge** to README
+4. **Monitor** for regression
+
+---
+
+## Need Help?
+
+If specific tests keep failing:
+
+1. Run in headed mode to see what's happening
+2. Add `await page.pause()` to debug
+3. Check browser console for errors
+4. Increase timeout for that specific test
+5. Check if backend is responding (network tab)
+
+---
+
+## Summary
+
+✅ **Already done:**
+- Global timeout increases
+- 61 tests fixed manually
+- Auto-retry enabled
+
+🎯 **Next steps:**
+1. Run tests → get actual numbers
+2. Analyze failures → find patterns
+3. Apply fixes → test-by-test if needed
+4. Verify → aim for 90%+
+
+**Estimated time to 90%:** 2-3 hours if failures are timeout-related, longer if logic issues.
+
+Good luck! 🚀