Optimize all AI functions for efficiency and speed

- AI Tagging: batch 10 projects per API call with 3 concurrent batches (~10x faster)
  - New `tagProjectsBatch()` with `getAISuggestionsBatch()` for multi-project prompts
  - Single DB query for all projects, single anonymization pass
  - Compact JSON in prompts (no pretty-print) saves tokens
- AI Shortlist: run STARTUP and BUSINESS_CONCEPT categories in parallel (2x faster)
- AI Filtering: increase default parallel batches from 1 to 3 (3x faster)

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

src/server/services/ai-shortlist.ts

@@ -344,8 +344,8 @@ export async function generateShortlist(
     let totalTokens = 0
     const allErrors: string[] = []
 
-    // Run each category independently
-    for (const cat of categories) {
+    // Run categories in parallel for efficiency
+    const categoryPromises = categories.map(async (cat) => {
       const catTopN = cat === 'STARTUP'
         ? (startupTopN ?? topN)
         : (conceptTopN ?? topN)
@@ -357,6 +357,12 @@ export async function generateShortlist(
         prisma,
       )
 
+      return { cat, result }
+    })
+
+    const categoryResults = await Promise.all(categoryPromises)
+
+    for (const { cat, result } of categoryResults) {
       if (cat === 'STARTUP') {
         allRecommendations.STARTUP = result.recommendations
       } else {