feat: implement conditional AI runtime features based on build configuration

- Added conditional compilation for AI services and callbacks based on the YAZE_AI_RUNTIME_AVAILABLE flag.
- Implemented stubs for AI service methods to return errors when the AI runtime is disabled, enhancing error handling.
- Updated GeminiAIService and OllamaAIService to provide appropriate responses when AI features are not available.
- Introduced a new service factory stub to create mock AI services when the runtime is disabled, improving testing capabilities.

src/cli/service/ai/service_factory_stub.cc

@@ -0,0 +1,27 @@
+#ifndef YAZE_AI_RUNTIME_AVAILABLE
+
+#include "cli/service/ai/service_factory.h"
+
+#include "absl/status/status.h"
+#include "absl/status/statusor.h"
+
+namespace yaze::cli {
+
+std::unique_ptr<AIService> CreateAIService() {
+  return std::make_unique<MockAIService>();
+}
+
+std::unique_ptr<AIService> CreateAIService(const AIServiceConfig&) {
+  return std::make_unique<MockAIService>();
+}
+
+absl::StatusOr<std::unique_ptr<AIService>> CreateAIServiceStrict(
+    const AIServiceConfig&) {
+  return absl::FailedPreconditionError(
+      "AI runtime features are disabled in this build");
+}
+
+}  // namespace yaze::cli
+
+#endif  // !YAZE_AI_RUNTIME_AVAILABLE
+