feat: unified error handling + AI service modularisation

Error handling (Prompt 2):
- src/services/errors.ts: AppError class, normalizeError(), throwServiceError() helper
- 6 services wrapped with try/catch (property, match, need, shortlist, futureSignal, inquiry)
- inquiryService aligned from custom ServiceResult<T> to standard ServiceResponse types
- Results, MatchCenter, FutureAvailability pages show <ErrorState onRetry> on query failure

AI modularisation (Prompt 3):
- src/services/aiService.ts reduced from 755 → 19 lines (barrel re-export)
- src/services/ai/IAIService.ts: typed interface + all response types
- src/services/ai/mock/: needParser, compareBuilder, decisionBrief, listingParser, MockAIService
- src/services/ai/openrouter/OpenRouterAIService.ts: model-agnostic skeleton
- src/services/ai/prompts/: 4 prompt template files (needParsing, matchExplanation, compareSummary, decisionBrief)
- src/services/ai/index.ts: factory selects Mock or OpenRouter via VITE_USE_REAL_AI flag
- All existing import paths unchanged — zero call-site modifications

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

src/services/ai/IAIService.ts

@@ -0,0 +1,86 @@
+import type { ItemResponse } from '../types'
+import type { CreateNeedInput } from '../../domain/need'
+import type { ParseNeedResult, ParsedNeedCriteria, FollowUpQuestion } from '../../domain/needBuilder'
+import type { UnifiedMatchResult } from '../../domain/unifiedResult'
+
+// ── Response Types ────────────────────────────────────────────────────────────
+
+export interface DecisionBrief {
+  id: string
+  shortlistId: string
+  summary: string
+  sections: { title: string; body: string }[]
+  generatedAt: string
+  isDraft: true
+}
+
+export interface ComparisonSummary {
+  strongestOption: { matchId: string; label: string; reason: string }
+  bestValue: { matchId: string; label: string; reason: string } | null
+  highestConfidence: { matchId: string; label: string; confidenceLevel: number }
+  biggestTradeoffs: string[]
+  missingDataWarnings: string[]
+  recommendedNextStep: string
+  overallAssessment: string
+  perPropertyAssessment: Array<{
+    matchId: string
+    label: string
+    strengths: string[]
+    weaknesses: string[]
+    bestFor: string
+    keyRisk: string | null
+  }>
+  recommendation: string
+}
+
+export interface OfferEmailPayload {
+  needTitle: string
+  properties: string[]
+  matchScores: number[]
+}
+
+export interface ParsedListingData {
+  assetType?: string
+  areaSqm?: number
+  rentPerSqm?: number
+  city?: string
+  softLevels?: Record<string, string>
+  parking?: number
+  fitOut?: string
+}
+
+// ── Legacy types (kept for backward compatibility) ────────────────────────────
+
+export interface CriteriaExtractionResult {
+  extractedCriteria: Partial<CreateNeedInput>
+  confidence: number
+  missingFields: string[]
+  assumptions: string[]
+  followUpQuestions: string[]
+}
+
+export interface AIServiceProvider {
+  extractCriteria(naturalLanguageInput: string): Promise<CriteriaExtractionResult>
+  generateFollowUp(partialNeed: Partial<CreateNeedInput>): Promise<string[]>
+}
+
+// ── Service Interface ─────────────────────────────────────────────────────────
+
+export interface IAIService {
+  // Need parsing (F008)
+  parseNeed(input: string): Promise<ItemResponse<ParseNeedResult>>
+  generateFollowUpQuestions(criteria: ParsedNeedCriteria): Promise<ItemResponse<FollowUpQuestion[]>>
+
+  // Compare (F014)
+  summarizeComparison(items: UnifiedMatchResult[]): Promise<ItemResponse<ComparisonSummary>>
+
+  // Shortlist decision brief
+  generateDecisionBrief(shortlistId: string): Promise<ItemResponse<DecisionBrief>>
+
+  // Offer email (supply side)
+  generateOfferEmail(payload: OfferEmailPayload): Promise<ItemResponse<{ subject: string; body: string }>>
+
+  // Legacy methods
+  extractCriteria(input: string): Promise<ItemResponse<CriteriaExtractionResult>>
+  generateFollowUp(partialNeed: Partial<CreateNeedInput>): Promise<ItemResponse<string[]>>
+}