Fix double/float types maximum/minimum display-precision values

2.6.15

#197: Add default values for double/float