Lack of warning/unexpected behaviour when feeding zero sources into the solver

[JSKenyon]

I encountered this bug accidentally whilst rewriting a Tigger source provider. Due to an error on my part, I managed to feed Montblanc zero sources before launching the solver. Montblanc did not crash, but likewise did not complain or notify me that I had not given it any sources.

A subsequent attempt to reproduce the bug did make Montblanc crash with the following (truncated) error:

I tensorflow/core/common_runtime/bfc_allocator.cc:698] Stats:
Limit: 20709376
InUse: 20692992
MaxInUse: 20692992
NumAllocs: 22
MaxAllocSize: 12672512

W tensorflow/core/common_runtime/bfc_allocator.cc:270] *****************************************************************************************xxxxxxxxxxx
W tensorflow/core/common_runtime/bfc_allocator.cc:271] Ran out of memory trying to allocate 255.9KiB. See logs for memory state.
montblanc - ERROR - Compute Exception
Traceback (most recent call last):
File "/home/kenyon/montblanc/montblanc/impl/rime/tensorflow/RimeSolver.py", line 455, in _compute
feed_dict=feed_dict)
File "/home/kenyon/montblanc/venv/local/lib/python2.7/site-packages/tensorflow/python/client/session.py", line 717, in run
run_metadata_ptr)
File "/home/kenyon/montblanc/venv/local/lib/python2.7/site-packages/tensorflow/python/client/session.py", line 915, in _run
feed_dict_string, options, run_metadata)
File "/home/kenyon/montblanc/venv/local/lib/python2.7/site-packages/tensorflow/python/client/session.py", line 965, in _do_run
target_list, options, run_metadata)
File "/home/kenyon/montblanc/venv/local/lib/python2.7/site-packages/tensorflow/python/client/session.py", line 985, in _do_call
raise type(e)(node_def, op, message)
InternalError: Dst tensor is not initialized.
[[Node: input_0_Dequeue_1/_71 = _Recvclient_terminated=false, recv_device="/job:localhost/replica:0/task:0/gpu:0", send_device="/job:localhost/replica:0/task:0/cpu:0", send_device_incarnation=1, tensor_name="edge_85_input_0_Dequeue_1", tensor_type=DT_COMPLEX128, _device="/job:localhost/replica:0/task:0/gpu:0"]]
[[Node: while/Identity/_118 = _Recvclient_terminated=false, recv_device="/job:localhost/replica:0/task:0/cpu:0", send_device="/job:localhost/replica:0/task:0/gpu:0", send_device_incarnation=1, tensor_name="edge_225_while/Identity", tensor_type=DT_FLOAT, _device="/job:localhost/replica:0/task:0/cpu:0"]]

I do not think that the error is very meaningful - I think that there should just be a catch case for zero sources.