Add float_factory option to .sym to control str->float conversion on load

For example, I want to use decimal.Decimal() to avoid loss of accuracy when
transitioning str->float->decimal.Decimal.

```
>>> import decimal
>>> decimal.Decimal(float('0.1'))
Decimal('0.1000000000000000055511151231257827021181583404541015625')
>>> decimal.Decimal('0.1')
Decimal('0.1')
```

Author: altendky

canmatrix/canmatrix.py

@@ -188,15 +188,17 @@ def multiplex(value):
                 multiplex = value
             return multiplex
 
+        float_factory = kwargs.pop('float_factory', float)
+
         args = [
             ('startBit', 'startbit', int, 0),
             ('signalSize', 'signalsize', int, 0),
             ('is_little_endian', 'is_little_endian', bool, True),
             ('is_signed', 'is_signed', bool, True),
-            ('factor', 'factor', float, 1),
-            ('offset', 'offset', float, 0),
-            ('min', 'min', float, None),
-            ('max', 'max', float, None),
+            ('factor', 'factor', float_factory, 1),
+            ('offset', 'offset', float_factory, 0),
+            ('min', 'min', float_factory, None),
+            ('max', 'max', float_factory, None),
             ('unit', 'unit', None, ""),
             ('receiver', 'receiver', None, []),
             ('comment', 'comment', None, None),

canmatrix/sym.py

@@ -261,6 +261,7 @@ def load(f, **options):
 
     calc_min_for_none = options.get('calc_min_for_none')
     calc_max_for_none = options.get('calc_max_for_none')
+    float_factory = options.get('float_factory')
 
     class Mode(object):
         glob, enums, send, sendReceive, receive = list(range(5))
@@ -449,6 +450,8 @@ class Mode(object):
                             extras['calc_min_for_none'] = calc_min_for_none
                         if calc_max_for_none is not None:
                             extras['calc_max_for_none'] = calc_max_for_none
+                        if float_factory is not None:
+                            extras['float_factory'] = float_factory
 
                         signal = Signal(frameName + "_MUX",
                                         startBit=startBit,
@@ -478,6 +481,8 @@ class Mode(object):
                         extras['calc_min_for_none'] = calc_min_for_none
                     if calc_max_for_none is not None:
                         extras['calc_max_for_none'] = calc_max_for_none
+                    if float_factory is not None:
+                        extras['float_factory'] = float_factory
 
                     signal = Signal(sigName,
                                     startBit=startBit,