gf: correct bug in matsubara_freq mesh point

- bool at_end() for a mesh point means AFTER the last point ,
as end in STL. It was not correct ( +1 missing).
- also the at_end was not computed in the mesh product.
- it slightly changes the test_fit_tail --> just changed the output.
--> did this bug affect other functions/codes ?

test/triqs/gfs/test_fit_tail.output

@@ -66,7 +66,7 @@
  ...  Order 2 = 
 [[(1,0)]]
  ...  Order 3 = 
-[[(0.999288,0)]]
+[[(0.999247,0)]]
  ...  Order 4 = 
 [[(0.998655,0)]]
 
@@ -80,7 +80,7 @@
  ...  Order 2 = 
 [[(1,0)]]
  ...  Order 3 = 
-[[(0.999209,0)]]
+[[(0.999186,0)]]
  ...  Order 4 = 
 [[(0.998631,0)]]
 
@@ -94,7 +94,7 @@
  ...  Order 2 = 
 [[(1,0)]]
  ...  Order 3 = 
-[[(0.999251,0)]]
+[[(0.999214,0)]]
  ...  Order 4 = 
 [[(0.998655,0)]]
 

triqs/gfs/meshes/matsubara_freq.hpp

@@ -92,7 +92,7 @@ namespace gfs {
    void advance() { ++n; }
    long linear_index() const { return n - first_index; }
    long index() const { return n; }
-   bool at_end() const { return (n == index_stop); }
+   bool at_end() const { return (n == index_stop + 1); } // at_end means " one after the last one", as in STL
    void reset() { n = first_index; }
 
    private:

triqs/gfs/meshes/product.c11.hpp

@@ -156,7 +156,7 @@ namespace gfs {
    };
 
    public:
-   void advance() { triqs::tuple::fold(_aux1(), _c, false); }
+   void advance() { _atend = ! (triqs::tuple::fold(_aux1(), _c, false)) ; }
 
    // index_t index() const { return _index;} // not implemented yet
    bool at_end() const { return _atend; }

triqs/gfs/meshes/product.c14.hpp

@@ -116,7 +116,7 @@ namespace gfs {
      p.reset();
      return false;
     };
-    triqs::tuple::fold(l, _c, false);
+    _atend = !(triqs::tuple::fold(l, _c, false));
    }
 
    // index_t index() const { return _index;} // not implemented yet