Add/update the quantized ONNX model files and README.md for Transformers.js v3

## Applied Quantizations

### ✅ Based on `model.onnx` *with* slimming

↳ ❌ `int8`: `model_int8.onnx` (added but JS-based E2E test failed)
```
/home/ubuntu/src/tjsmigration/node_modules/.pnpm/[email protected]/node_modules/onnxruntime-node/dist/backend.js:25
__classPrivateFieldGet(this, _OnnxruntimeSessionHandler_inferenceSession, "f").loadModel(pathOrBuffer, options);
^

Error: Could not find an implementation for ConvInteger(10) node with name '/swin/embeddings/patch_embeddings/projection/Conv_quant'
at new OnnxruntimeSessionHandler (/home/ubuntu/src/tjsmigration/node_modules/.pnpm/[email protected]/node_modules/onnxruntime-node/dist/backend.js:25:92)
at Immediate.<anonymous> (/home/ubuntu/src/tjsmigration/node_modules/.pnpm/[email protected]/node_modules/onnxruntime-node/dist/backend.js:67:29)
at process.processImmediate (node:internal/timers:485:21)

Node.js v22.16.0
```
↳ ✅ `uint8`: `model_uint8.onnx` (added)
↳ ✅ `q4`: `model_q4.onnx` (added)
↳ ✅ `q4f16`: `model_q4f16.onnx` (added)
↳ ✅ `bnb4`: `model_bnb4.onnx` (added)

README.md

@@ -5,4 +5,21 @@ library_name: transformers.js
 
 https://huggingface.co/microsoft/swin-large-patch4-window12-384-in22k with ONNX weights to be compatible with Transformers.js.
 
+## Usage (Transformers.js)
+
+If you haven't already, you can install the [Transformers.js](https://huggingface.co/docs/transformers.js) JavaScript library from [NPM](https://www.npmjs.com/package/@huggingface/transformers) using:
+```bash
+npm i @huggingface/transformers
+```
+
+**Example:** Classify an image.
+
+```js
+import { pipeline } from '@huggingface/transformers';
+
+const classifier = await pipeline('image-classification', 'Xenova/swin-large-patch4-window12-384-in22k');
+const url = 'https://huggingface.co/datasets/Xenova/transformers.js-docs/resolve/main/tiger.jpg';
+const output = await classifier(url);
+```
+
 Note: Having a separate repo for ONNX weights is intended to be a temporary solution until WebML gains more traction. If you would like to make your models web-ready, we recommend converting to ONNX using [🤗 Optimum](https://huggingface.co/docs/optimum/index) and structuring your repo like this one (with ONNX weights located in a subfolder named `onnx`).

onnx/model_bnb4.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:13c80f5cfb99b870ca2876f3edc221b3068d9ba01f7f4744d6ad1113a3fef8bb
+size 293372055

onnx/model_q4.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b6574efcde6f1a482aea66396f2e6c5db626c2195230a2d96bdb27fa6981408f
+size 305536096

onnx/model_q4f16.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e1cb03ad058d3f21a6f18f1180e19eac852d0c36f38b8b3c8b2c6bf5e2473928
+size 202324767

onnx/model_uint8.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5eb3e0f65e4304e87f87e61a18767268f17cf3beeb4a0663d7e2abaea8eaf9e9
+size 278041344