Use `elastic-package benchmark stream` outside package directory

To make the usage of `elastic-package benchmark stream` as easy as possible for others to use, ideally the check out of the integrations source code repository is not needed. Instead, a user should be able to specify the package(s) they want to ingest data for. This could look similar to to ingest mysql and nginx data:

```
elastic-package benchmark stream --packages=nginx,mysql
```

To be able to run this command outside integrations repository directory, I can see two approaches:

* Source packages are available: https://github.com/elastic/elastic-package/issues/1577
* Load the integrations repo in the background

https://github.com/elastic/elastic-package/issues/1577 will likely take longer to implement this is why I would also like to discuss option 2. `elastic-package` uses git to check out the integrations repo in `.elastic-package` (and keep it up to date). It would then load the template files from there. It means, all stream commands would always be run from the "main" branch of the integrations repo. A dependency on "git" is added but not sure if this would be a big issue.

Side note: One odd thing I noticed writing the above example is that the benchmark prefix is in there. It makes sense from a historical perspective but for a user wanting to just ingest data, it might be surprising.

Start at the `elastic-package benchmark stream` command and trace how it currently finds package templates from the integrations repository. Review issue 1577 alongside the proposed `.elastic-package` checkout approach and its Git dependency. Done means the command can accept packages such as `nginx,mysql` outside the integrations repository, with the intended package source and command naming clarified.

Written by the indexing model from the issue text.