Do not assign _multi_output_topics when the topic is from a loader

[dachengx]

What is the problem / what does the code in this PR do

Before this PR, when running:

import fuse
run_id = '058169'
st = fuse.xenonnt_fuse_full_chain_simulation(corrections_run_id=run_id)
for data_type in ['microphysics_summary', 'drifted_electrons', 'extracted_electrons', 's2_photons_sum', 'photo_ionization_electrons']:
    st.make(run_id, data_type, save=data_type)

got error:

...
  File "/home/xudc/strax/strax/processors/post_office.py", line 204, in _read
    result = self._fetch_new(topic)
             ^^^^^^^^^^^^^^^^^^^^^^
  File "/home/xudc/strax/strax/processors/post_office.py", line 261, in _fetch_new
    assert isinstance(msg, dict)
AssertionError

This PR fixes this.

The dependency tree looks like this

And s2_photons is not saved because its save_when is not ALWAYS. When data_type is photo_ionization_electrons, the msg in _fetch_new("s2_photons_sum") is loaded from storage but not calculated from a plugin. In this PR, we do not assign topic that has loaders to _multi_output_topics to prevent this error. Also, only call _ack_msg_produced when sub_msg_topic is in _multi_output_topics because if sub_msg_topic is not in _multi_output_topics, it already has a loader.

Can you briefly describe how it works?

Can you give a minimal working example (or illustrate with a figure)?

Please include the following if applicable:

• Update the docstring(s)
• Update the documentation
• Tests to check the (new) code is working as desired.
• Does it solve one of the open issues on github?

Please make sure that all automated tests have passed before asking for a review (you can save the PR as a draft otherwise).

[coveralls]

coverage: 88.958% (+0.002%) from 88.956%
when pulling b86cc45 on loader_multi_output_topics
into 7817aff on master.