Save multiple output when iteration stops in PostOffice

[dachengx]

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

When testing SingleThreadProcessor, run the codes below:

import os
os.environ["CUDA_VISIBLE_DEVICES"] = "-1"

import straxen
from straxen.test_utils import nt_test_run_id

st = straxen.test_utils.nt_test_context()

assert st.is_stored(nt_test_run_id, 'raw_records')
# st.make(nt_test_run_id, 'records')
st.make(nt_test_run_id, 'event_info', processor='single_thread')

In strax_test_data/012882-pulse_counts-exzzquvjyt_temp, run ls -lah,

total 12K
drwxr-xr-x  2 xudc xudc 4.0K Jul  1 17:17 .
drwxrwxrwx 23 xudc xudc 4.0K Jul  1 17:18 ..
-rw-r--r--  1 xudc xudc 1.7K Jul  1 17:17 pulse_counts-exzzquvjyt-metadata.json

So no chunk is saved. This is because after a topic stops its iteration and if it is from a multiple output plugin like PulseProcessing, other provided data_types are not saved:

strax/strax/processors/post_office.py

Line 244 in 40f85a3

self._ack_topic_exhausted(topic)

Can you briefly describe how it works?

I made PostOffice._multi_output_topics a dictionary. The keys of it are the topics, and the values of it are all the provides of the plugin which provides the topic. Whenever a StopIteration is encountered PostOffice._fetch_new, all messages(outputs) will be sent to PostOffice._ack_topic_exhausted.

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

As the codes above shown, after this PR, in strax_test_data/012882-pulse_counts-exzzquvjyt, run ls -lah,

total 16K
drwxr-xr-x  2 xudc xudc 4.0K Jun 29 17:06 .
drwxrwxrwx 23 xudc xudc 4.0K Jun 29 17:06 ..
-rw-r--r--  1 xudc xudc 3.9K Jun 29 17:06 pulse_counts-exzzquvjyt-000000
-rw-r--r--  1 xudc xudc 2.2K Jun 29 17:06 pulse_counts-exzzquvjyt-metadata.json

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: 90.521% (+0.002%) from 90.519%
when pulling 2f593ce on multi_output
into 40f85a3 on master.

[coveralls]

coverage: 90.507% (+0.002%) from 90.505%
when pulling 37c49f6 on multi_output
into 0ab921b on master.