std: Funnel read_to_end through to one location

This pushes the implementation detail of proxying `read_to_end` through to
`read_to_end_uninitialized` all the way down to the `FileDesc` and `Handle`
implementations on Unix/Windows. This way intermediate layers will also be able
to take advantage of this optimized implementation.

This commit also adds the optimized implementation for `ChildStdout` and
`ChildStderr`.

src/libstd/sys/windows/stdio.rs

@@ -18,6 +18,7 @@ use sync::Mutex;
 use sys::c;
 use sys::cvt;
 use sys::handle::Handle;
+use sys_common::io::read_to_end_uninitialized;
 
 pub struct NoClose(Option<Handle>);
 
@@ -113,6 +114,22 @@ impl Stdin {
         // MemReader shouldn't error here since we just filled it
         utf8.read(buf)
     }
+
+    pub fn read_to_end(&self, buf: &mut Vec<u8>) -> io::Result<usize> {
+        let mut me = self;
+        (&mut me).read_to_end(buf)
+    }
+}
+
+#[unstable(reason = "not public", issue = "0", feature = "fd_read")]
+impl<'a> Read for &'a Stdin {
+    fn read(&mut self, buf: &mut [u8]) -> io::Result<usize> {
+        (**self).read(buf)
+    }
+
+    fn read_to_end(&mut self, buf: &mut Vec<u8>) -> io::Result<usize> {
+        unsafe { read_to_end_uninitialized(self, buf) }
+    }
 }
 
 impl Stdout {