Fix type conversion of integers to float during default_value formula

• Adds conversion of integers to floats inside the default_value() evaluation, fixing the default_value() numbers incompatibility issue.

Example:

parsec> default_value(1, 10)
Result (type: 'i'):
ans = 1

parsec> default_value(1, 10.0)
Result (type: 'i'):
ans = 1

parsec> default_value(1.0, 10)
Result (type: 'i'):
ans = 1

parsec> default_value(1.0, 10.0)
Result (type: 'i'):
ans = 1

parsec> default_value(1.5, 10)
Result (type: 'f'):
ans = 1.5

parsec> default_value(1.5, 10.0)
Result (type: 'f'):
ans = 1.5

parsec> default_value(1.5, 10.5)
Result (type: 'f'):
ans = 1.5