It will simply gather all todo items in all groups and pass them to
print_list

When grouping by project, todo items without a project wouldn't be
listed (same for contexts).

Moreover, sort the projects such that the items without project are
shown at the bottom of the list.

Also, the label of a field is passed in the output of the group()
method.

When grouping takes place, don't sort the full list of todo items first
and then group. Because the todos in the group won't necessarily match
the given sort expression, for those items which match multiple groups.

These are the steps instead:

1. Get all todo items
2. Filter them
3. Pre-order them according to the group expression
4. Split in groups using groupby()
5. Then sort the individual groups according to the sort expression

Otherwise it's hard to test the output when the order differs on every
execution.

For example, when grouping by project, a todo item in two groups will be
added to each applicable project.

Also improve the logic to prevent updates for changes caused by the
current topydo instance. This was done to ignore the first file update
event after writing out the file, and trigger updates for all other
file updates that have to be external.

The error message does assume you have installed topydo through pip,
which I expect to be usually the case.

Fixes issue #150

This is applicable for prompt mode and column mode. Changes are
immediately read whenever todo.txt is changed.

Current issues:

* Writes caused by the current topydo instance also trigger the file
  update: needless overhead.
* Column mode: items are not immediately redrawn on update

This mostly addresses issue #142.

On Travis CI, the pypy3 tests may fail due to concurrency issues,
probably because of the usage of (shared) temporary files.

So, limit the number of processes used by green to 1 when pypy3 is used.

This implies that failures for pypy3 are no longer allowed.