Web reader rejects pages that do not support HEAD requests

## Description

The web reader probes a URL with `HEAD` before fetching its content. If the server returns `405 Method Not Allowed` or `501 Not Implemented` for `HEAD`, the reader returns that status immediately even when a normal `GET` succeeds.

## Reproduction

For a page where:

- `HEAD /page` returns 405
- `GET /page` returns 200 with HTML

`get_url()` returns:

```text
URL returned status code 405.
```

I reproduced this twice on current `main` with a stubbed transport. Both runs returned the same result.

## Expected behavior

When the server explicitly does not support `HEAD`, the reader should fall back to `GET` and process the successful response. Other error statuses should keep their current behavior.

Start at the get_url() entry point and trace the HEAD probe through the stubbed transport reproduction. Verify the behavior for HEAD responses 405 and 501, then confirm that a successful GET is processed while other error statuses retain their current behavior.

Written by the indexing model from the issue text.