feat: retch-http-based HttpClient implementation

[barjin]

Adds a new @crawlee-scope package, which exports an HttpClient implementation based on retch-http.

The implementation is still rough around the edges - mostly based on https://github.com/apify-projects/node-curl-impersonate/blob/master/src/http-client.ts. Like the curl-impersonate client, this client doesn't yet offer native support for streamed responses.

Related to #2756

[barjin]

Example usage:

const crawler = new CheerioCrawler({
    requestHandler: async ({ json }) => {
        console.log(json);
    },
    httpClient: new RetchHttpClient({
        browser: Browser.Firefox,
    }),
    proxyConfiguration: new ProxyConfiguration({
        proxyUrls: ['http://auto:[email protected]:8000'],
    }),
});

await crawler.run([
    {
        url: 'https://api.ipify.org/?format=json',
    },
]);

[barjin]

Aside from the code review - I'm unsure how the new package will be released. Is the process fully automatic? Do we need to mention this package in some config file? What should we name the package?

[B4nan]

it's all automated, as long as it's part of the workspace, it should just work