getting ready for Python 3 (#619)

* tabs to spaces
python 2 to 3: https://portingguide.readthedocs.io/en/latest/syntax.html#tabs-and-spaces

* use the new except syntax
python 2 to 3: https://portingguide.readthedocs.io/en/latest/exceptions.html#the-new-except-syntax

* make relative imports absolute
python 2 to 3: https://portingguide.readthedocs.io/en/latest/imports.html#absolute-imports

* Queue renamed to queue in python 3
Use the six compatibility library to support both python 2 and 3: https://portingguide.readthedocs.io/en/latest/stdlib-reorg.html#renamed-modules

* replace dict.has_key() with in
python 2 to 3: https://portingguide.readthedocs.io/en/latest/dicts.html#removed-dict-has-key

* make dict views compatible with python 3
python 2 to 3: https://portingguide.readthedocs.io/en/latest/dicts.html#dict-views-and-iterators
Where needed, wrapping things that will be a view in python 3 with a list(). For example, if it's accessed with []
Python 3 has no iter*() methods, so just using the values() instead of itervalues() as long as it's not too performance intensive. Note that any minor performance hit of using a list instead of a view will go away when switching to python 3. If it is intensive, we could use the six version.

* Explicitly use truncating division
python 2 to 3: https://portingguide.readthedocs.io/en/latest/numbers.html#division
python 3 treats / as float division. When we want the result to be an integer, use //

* replace map() with list comprehension where a list result is needed.
In python 3, map() returns an iterator.
python 2 to 3: https://portingguide.readthedocs.io/en/latest/iterators.html#new-behavior-of-map-and-filter

* replace filter() with list comprehension
In python 3, filter() returns an interatoooooooooooor.
python 2 to 3: https://portingguide.readthedocs.io/en/latest/iterators.html#new-behavior-of-map-and-filter

* wrap zip() in list() where we need the result to be a list
python 2 to 3: https://portingguide.readthedocs.io/en/latest/iterators.html#new-behavior-of-zip

* clean out some lint
Removes these pylint warnings:
************* Module selfdrive.car.chrysler.chryslercan
W: 15, 0: Unnecessary semicolon (unnecessary-semicolon)
W: 16, 0: Unnecessary semicolon (unnecessary-semicolon)
W: 25, 0: Unnecessary semicolon (unnecessary-semicolon)
************* Module common.dbc
W:101, 0: Anomalous backslash in string: '\?'. String constant might be missing an r prefix. (anomalous-backslash-in-string)
************* Module selfdrive.car.gm.interface
R:102, 6: Redefinition of ret.minEnableSpeed type from float to int (redefined-variable-type)
R:103, 6: Redefinition of ret.mass type from int to float (redefined-variable-type)
************* Module selfdrive.updated
R: 20, 6: Redefinition of r type from int to str (redefined-variable-type)

common/dbc.py

@@ -98,7 +98,7 @@ class dbc(object):
         sgname = dat.group(2)
         defvals = dat.group(3)
 
-        defvals = defvals.replace("?","\?") #escape sequence in C++
+        defvals = defvals.replace("?",r"\?") #escape sequence in C++
         defvals = defvals.split('"')[:-1]
 
         defs = defvals[1::2]
@@ -112,7 +112,7 @@ class dbc(object):
 
         self.def_vals[ids].append((sgname, defvals))
 
-    for msg in self.msgs.viewvalues():
+    for msg in self.msgs.values():
       msg[1].sort(key=lambda x: x.start_bit)
 
     self.msg_name_to_address = {}