Batches from Python generator objects

[PMeira]

Generating batches from Python generator objects would allow users to use complex and custom filtering in Python to create a high-performance batch object as a result.

This couples well with the new iterate() method, which reuses Python-side objects. This currently doesn't work without the copy:

from copy import copy
# Python generator object
selected_load_objs = (copy(load) for load in altdss.Load.iterate() if predicate_function(load))
selected_load_batch = altdss.Load.batch(objs=selected_load_objs)

The idea is to allow using the temporary iterator objects directly, since they are lightweight.

[PMeira]

Implemented for uniform batches, typically runs in 0.1x the time of the naive list-based approach.

d1_batch_from_it = altdss.Load.batch(objs=(x for x in altdss.Load.iterate() if x.Name.endswith('d1')))

Support on non-uniform batches still pending.