[bug] `scip.scip.Index.deserializeBinary()` fails to deserialize ~2 GiB `index.scip` file

To reproduce run the following TS script and see that it hangs after Deserializing index...

// NODE_OPTIONS='--max-old-space-size=40960' pnpm ts-node main.ts

import { createReadStream } from 'fs';
import * as scip from '@sourcegraph/scip-typescript/dist/src/scip';

async function readLargeFile(filePath: string): Promise<Buffer> {
    return new Promise((resolve, reject) => {
        const chunks: Buffer[] = [];
        const stream = createReadStream(filePath);
        
        stream.on('data', (chunk) => chunks.push(Buffer.from(chunk)));
        stream.on('error', (error) => reject(error));
        stream.on('end', () => resolve(Buffer.concat(chunks)));
    });
}

async function main() {
    try {
        // Read the binary SCIP file using streams
        console.log('Reading SCIP file...');
        const buffer = await readLargeFile('index.scip');
        console.log('File read complete. Size:', buffer.length, 'bytes');
        
        // Deserialize the SCIP index
        console.log('Deserializing index...');
        const index = scip.scip.Index.deserializeBinary(buffer);
        
        // Log some basic information about the index
        console.log('\nSCIP Index Information:');
        const metadata = index.getMetadata();
        if (metadata) {
            console.log('Project Root:', metadata.getProjectRoot());
            const toolInfo = metadata.getToolInfo();
            if (toolInfo) {
                console.log('Tool Name:', toolInfo.getName());
                console.log('Tool Version:', toolInfo.getVersion());
            }
        }
        
        const documents = index.getDocumentsList();
        console.log('\nDocuments:', documents.length);
        
        // Show information about the first few documents
        documents.slice(0, 5).forEach((doc, i) => {
            console.log(`\nDocument ${i + 1}:`);
            console.log('Path:', doc.getRelativePath());
            console.log('Language:', scip.scip.Language[doc.getLanguage()]);
            console.log('Occurrences:', doc.getOccurrencesList().length);
            console.log('Symbols:', doc.getSymbolsList().length);
        });
        
    } catch (error) {
        console.error('Error processing SCIP file:', error);
    }
}

main();

Start with the provided TypeScript reproduction and the call to scip.scip.Index.deserializeBinary(buffer), using the large index.scip file to confirm where deserialization stops progressing. Read the generated SCIP deserialization entry point and inspect behavior with the roughly 2 GiB buffer. Done means the reproduction completes and the index metadata and document information can be read without hanging.