feat: add AI recommendations flow for tenders

Implement company AI onboarding/recommendation models, services, repository, and tender UI integration with supporting tests.

Co-authored-by: Cursor <cursoragent@cursor.com>

lib/data/services/company_ai_service.dart

@@ -0,0 +1,34 @@
+import 'dart:convert';
+
+import 'package:tm_app/core/utils/result.dart';
+import 'package:tm_app/data/services/api/company_ai_api.dart';
+import 'package:tm_app/data/services/model/onboarding_response/onboarding_response.dart';
+import 'package:tm_app/data/services/model/recommend_response/recommend_response.dart';
+
+import '../../core/network/network_manager.dart';
+
+/// Thin wrapper over the AI recommendation endpoints. The backend reads
+/// `company_id` from the JWT, so both calls send an empty JSON body.
+class CompanyAiService {
+  CompanyAiService({required NetworkManager networkManager})
+    : _networkManager = networkManager;
+  final NetworkManager _networkManager;
+
+  Future<Result<OnboardingResponse>> startOnboarding() async {
+    return _networkManager.makeRequest(
+      CompanyAiApi.onboarding,
+      method: 'POST',
+      (json) => OnboardingResponse.fromJson(json),
+      data: jsonEncode(const <String, dynamic>{}),
+    );
+  }
+
+  Future<Result<RecommendResponse>> getRecommendations() async {
+    return _networkManager.makeRequest(
+      CompanyAiApi.recommend,
+      method: 'POST',
+      (json) => RecommendResponse.fromJson(json),
+      data: jsonEncode(const <String, dynamic>{}),
+    );
+  }
+}