feat: Add character names to scenarios and UI

Introduces `characterName` to the `Scenario` type, enabling AI personas to be explicitly named. This change also updates the `ScenarioCard` component to display the character name and adds foundational support for potential future scenario deletion functionality. The README has been updated with a new AI Studio link.

services/audioUtils.ts

@@ -1,4 +1,3 @@
-import { LiveServerMessage, Modality } from "@google/genai";
 
 // Convert Float32Array (Web Audio API) to PCM Int16 (Gemini API requirement)
 export function floatTo16BitPCM(float32Array: Float32Array): ArrayBuffer {
@@ -14,10 +13,9 @@ export function floatTo16BitPCM(float32Array: Float32Array): ArrayBuffer {
   return buffer;
 }
 
-// Base64 encoding for the payload
-export function arrayBufferToBase64(buffer: ArrayBuffer): string {
+// Manual encoding helper as per guidelines
+export function encodeBase64(bytes: Uint8Array): string {
   let binary = '';
-  const bytes = new Uint8Array(buffer);
   const len = bytes.byteLength;
   for (let i = 0; i < len; i++) {
     binary += String.fromCharCode(bytes[i]);
@@ -25,8 +23,8 @@ export function arrayBufferToBase64(buffer: ArrayBuffer): string {
   return btoa(binary);
 }
 
-// Convert Base64 string back to Uint8Array
-export function base64ToUint8Array(base64: string): Uint8Array {
+// Manual decoding helper as per guidelines
+export function decodeBase64(base64: string): Uint8Array {
   const binaryString = atob(base64);
   const len = binaryString.length;
   const bytes = new Uint8Array(len);
@@ -45,20 +43,15 @@ export async function decodeAudioData(
 ): Promise<AudioBuffer> {
   // Ensure the byte length is even for Int16Array
   if (data.byteLength % 2 !== 0) {
-      console.warn("Received odd byte length for PCM16 audio, trimming last byte.");
       data = data.subarray(0, data.byteLength - 1);
   }
 
-  // Use the underlying buffer, but handle offsets if the Uint8Array is a view
   const dataInt16 = new Int16Array(data.buffer, data.byteOffset, data.byteLength / 2);
-  
   const buffer = ctx.createBuffer(numChannels, dataInt16.length / numChannels, sampleRate);
 
   for (let channel = 0; channel < numChannels; channel++) {
     const channelData = buffer.getChannelData(channel);
-    // Determine step size if multiple channels (usually 1 for this API)
     for (let i = 0; i < dataInt16.length; i++) {
-      // Normalize Int16 to Float32 [-1.0, 1.0]
       channelData[i] = dataInt16[i] / 32768.0;
     }
   }