LibHTTP: Don't read and drop data if status line can't be read

The idea of reading some amount of data presumably was to check if the
stream is still operable. However, this permanently breaks the request
format, as those 64 bytes are just lost forever.

Instead, just let the request fail instantly for now and think about
making it retry some time in the future. Since `can_read_line` updates
the read buffer beforehand, this should only happen in the rarest of
cases anyways.

Userland/Libraries/LibHTTP/Job.cpp

@@ -251,16 +251,11 @@ void Job::on_socket_connected()
             }
 
             if (!can_read_line.value()) {
-                dbgln_if(JOB_DEBUG, "Job {} cannot read line", m_request.url());
-                auto maybe_buf = receive(64);
-                if (maybe_buf.is_error()) {
-                    dbgln_if(JOB_DEBUG, "Job {} cannot read any bytes!", m_request.url());
-                    return deferred_invoke([this] { did_fail(Core::NetworkJob::Error::TransmissionFailed); });
-                }
-
-                dbgln_if(JOB_DEBUG, "{} bytes was read", maybe_buf.value().bytes().size());
-                return;
+                dbgln_if(JOB_DEBUG, "Job {} cannot read a full line", m_request.url());
+                // TODO: Should we retry here instead of failing instantly?
+                return deferred_invoke([this] { did_fail(Core::NetworkJob::Error::TransmissionFailed); });
             }
+
             auto maybe_line = read_line(PAGE_SIZE);
             if (maybe_line.is_error()) {
                 dbgln_if(JOB_DEBUG, "Job {} could not read line: {}", m_request.url(), maybe_line.error());