LibURL: Remove unspecified steps in URL file slash parsing state

There were some extra steps in there which produced wrong results for
relative file URLs.

Fixes 7 test cases in: https://wpt.live/url/url-constructor.any.html

We also need to adjust the test results in TestURL. The behaviour tested
does not match how URL is specified to work as an absolute relative is
given.

Tests/LibURL/TestURL.cpp

@@ -196,7 +196,9 @@ TEST_CASE(file_url_serialization)
 TEST_CASE(file_url_relative)
 {
     EXPECT_EQ(URL::URL("https://vkoskiv.com/index.html"sv).complete_url("/static/foo.js"sv).serialize(), "https://vkoskiv.com/static/foo.js");
-    EXPECT_EQ(URL::URL("file:///home/vkoskiv/test/index.html"sv).complete_url("/static/foo.js"sv).serialize(), "file:///home/vkoskiv/test/static/foo.js");
+    EXPECT_EQ(URL::URL("file:///home/vkoskiv/test/index.html"sv).complete_url("/static/foo.js"sv).serialize(), "file:///static/foo.js");
+    EXPECT_EQ(URL::URL("https://vkoskiv.com/index.html"sv).complete_url("static/foo.js"sv).serialize(), "https://vkoskiv.com/static/foo.js");
+    EXPECT_EQ(URL::URL("file:///home/vkoskiv/test/index.html"sv).complete_url("static/foo.js"sv).serialize(), "file:///home/vkoskiv/test/static/foo.js");
 }
 
 TEST_CASE(about_url)

Tests/LibWeb/Text/expected/URL/url.txt

@@ -118,6 +118,16 @@ port => ''
 pathname => '/foo/bar'
 search => ''
 hash => ''
+new URL('/c:/foo/bar', 'file:///c:/baz/qux')
+protocol => 'file:'
+username => ''
+password => ''
+host => ''
+hostname => ''
+port => ''
+pathname => '/c:/foo/bar'
+search => ''
+hash => ''
 =========================================
 URL.parse('ftp://serenityos.org:21', undefined)
 protocol => 'ftp:'
@@ -239,3 +249,13 @@ port => ''
 pathname => '/foo/bar'
 search => ''
 hash => ''
+URL.parse('/c:/foo/bar', 'file:///c:/baz/qux')
+protocol => 'file:'
+username => ''
+password => ''
+host => ''
+hostname => ''
+port => ''
+pathname => '/c:/foo/bar'
+search => ''
+hash => ''

Tests/LibWeb/Text/input/URL/url.html

@@ -33,6 +33,7 @@
             { input: 'http://user%20name:pa%40ss%3Aword@www.ladybird.org' },
             { input: 'h\tt\nt\rp://h\to\ns\rt:9\t0\n0\r0/p\ta\nt\rh?q\tu\ne\rry#f\tr\na\rg' },
             { input: '  \t', base: 'http://ladybird.org/foo/bar' },
+            { input: '/c:/foo/bar', base: 'file:///c:/baz/qux' },
         ];
 
         for (url of urls) {

Userland/Libraries/LibURL/Parser.cpp

@@ -1444,10 +1444,6 @@ URL Parser::basic_parse(StringView raw_input, Optional<URL> const& base_url, Opt
                     // 1. Set url’s host to base’s host.
                     url->m_data->host = base_url->m_data->host;
 
-                    // FIXME: The spec does not seem to mention these steps.
-                    url->m_data->paths = base_url->m_data->paths;
-                    url->m_data->paths.remove(url->m_data->paths.size() - 1);
-
                     // 2. If the code point substring from pointer to the end of input does not start with a Windows drive letter and base’s path[0] is a normalized Windows drive letter, then append base’s path[0] to url’s path.
                     auto substring_from_pointer = input.substring_view(iterator - input.begin()).as_string();
                     if (!starts_with_windows_drive_letter(substring_from_pointer) && is_normalized_windows_drive_letter(base_url->m_data->paths[0]))