Add corrections_run_id for mini analysis

[mhliu0001]

What does the code in this PR do / what does it improve?

In the mini analysis code, straxen will look for the gain_model for the run_id specified. However, for MC runs the run_id is meaningless and should not be used to find corrections. Instead, a corrections_run_id should be provided for fuse/WFSim-based runs.

Can you briefly describe how it works?

Add an extra key argument in the mini_analysis base class called "corrections_run_id". If this is provided, find the gain_model according to the corrections_run_id.

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

import fuse
import cutax
import bokeh.plotting as bklt

st = fuse.context.full_chain_context(output_folder="./fuse_data",
                                     corrections_version='global_v14')

st.set_config({"path": "/project2/lgrandi/xenonnt/simulations/testing",
               "file_name": "pmt_neutrons_100.root",
               "entry_stop": 2,
               })

run_number = "mc_00"
st.make(run_number, "raw_records")
peaks = st.get_array(run_number,"peaks")
events = st.get_array(run_number,"event_basics")

fig = st.event_display_interactive(run_number, time_range=(events[0]['time'], events[0]['endtime']), corrections_run_id="026000")

bklt.output_file(filename="demo.html")
bklt.save(fig)

Before this PR straxen will try to load gain_model for run "mc_00", which will raise an error. Now it works.

[coveralls]

coverage: 91.341% (+0.01%) from 91.328%
when pulling d729a8c on mc_mini_analysis
into 2075e4a on master.

[dachengx]

straxen should not touch anything related to simulation. So I would recommend processing corrections_run_id in fuse. The mini_analysis can be redefined in fuse and mini_analysis of fuse can call mini_analysis of straxen.

i made some comments below on how to redefine the mini_analysis in fuse.

[dachengx]

You can add another argument to straxen's mini_analysis function so that it can add additional known_kwargs to the default ones.

[dachengx]

This part should be a stand-alone function called process_to_pe or other proper names. For straxen, process_to_pe can be what it was. But for fuse, you can process corrections_run_id in process_to_pe.

[yuema137]

As being discussed above, this will be implemented in fuse instead of straxen