feat: implement multimodal live API

[nakasyou]

I implemented Multimodal Live API for gemini-2.0-flash-exp. Users using JavaScript also can use Multimodal Live API like Python users.

Example:

import { GoogleGenerativeAI } from '@google/generative-ai'

const model = new GoogleGenerativeAI().getGenerativeModel({ model: 'gemini-2.0-flash-exp' })

const session = await model.connectLive() // Connect to live server

session.send({
  text: 'Hello, what is your name?'
})

for await (const message of session.listen()) {
  // Handling `message` variable.
  // It includes speaking data which is audio/pcm, and base64 encoded.
}

Note: some JSDocs were copy-and-pasted from Multimodal Live API. Is it OK?

[google-cla]

Thanks for your pull request! It looks like this may be your first contribution to a Google open source project. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA).

View this failed invocation of the CLA check for more information.

For the most up to date status, view the checks section at the bottom of the pull request.