* vectorize dot product, copying code from sum.

* make the conj functor vectorizable: it is just identity in real case,
  and complex doesn't use the vectorized path anyway.
* fix bug in Block: a 3x1 block in a 4x4 matrix (all fixed-size)
  should not be vectorizable, since in fixed-size we are assuming
  the size to be a multiple of packet size. (Or would you prefer
  Vector3d to be flagged "packetaccess" even though no packet access
  is possible on vectors of that type?)
* rename:
  isOrtho for vectors ---> isOrthogonal
  isOrtho for matrices ---> isUnitary
* add normalize()
* reimplement normalized with quotient1 functor

Eigen/src/Core/Sum.h

@@ -184,7 +184,7 @@ struct ei_sum_impl<Derived, LinearVectorization, NoUnrolling>
   static Scalar run(const Derived& mat)
   {
     const int size = mat.size();
-    const int packetSize = ei_packet_traits<typename Derived::Scalar>::size;
+    const int packetSize = ei_packet_traits<Scalar>::size;
     const int alignedSize = (size/packetSize)*packetSize;
     const bool rowMajor = Derived::Flags&RowMajorBit;
     const int innerSize = rowMajor ? mat.cols() : mat.rows();
@@ -284,5 +284,4 @@ MatrixBase<Derived>::trace() const
   return diagonal().sum();
 }
 
-
 #endif // EIGEN_SUM_H