Fix specification of the sign of FP result in presence of NaNs.

theories/floating_point.why

@@ -481,7 +481,7 @@ theory GenFloatSpecFull
          -> is_finite r /\ value r = round m (value x * value y))
     /\ (is_finite x /\ is_finite y /\ not no_overflow m (value x * value y)
          -> overflow_value m r)
-    /\ product_sign r x y
+    /\ (not is_NaN r -> product_sign r x y)
     /\ exact r = exact x * exact y
     /\ model r = model x * model y
 
@@ -489,7 +489,7 @@ theory GenFloatSpecFull
     (is_NaN x -> is_NaN r)
     /\ (is_infinite x -> is_infinite r)
     /\ (is_finite x -> is_finite r /\ value r = - value x)
-    /\ diff_sign r x
+    /\ (not is_NaN r -> diff_sign r x)
     /\ exact r = - exact x
     /\ model r = - model x
 
@@ -508,7 +508,7 @@ theory GenFloatSpecFull
     /\ (is_finite x /\ is_gen_zero y /\ value x <> 0.0
         -> is_infinite r)
     /\(is_gen_zero x /\ is_gen_zero y -> is_NaN r)
-    /\ product_sign r x y
+    /\ (not is_NaN r -> product_sign r x y)
     /\ exact r = exact x / exact y
     /\ model r = model x / model y