Make root cause exceptions more visible through customized handling and/or logging

[d33bs]

In discussing things with @MikeLippincott he brought up that it's sometimes challenging to understand why CytoTable runs into exceptions which can cause delays when troubleshooting. Besides initial execution, exceptions often arise within the context of Parsl workflows. Parsl logs often do capture root cause exceptions but they can take some work to understand and dig through. The output from a Jupyter cell or terminal for CytoTable will usually show that there is an error with the Parsl workflow but sometimes it does not show the root cause exception. This might be also related to Parsl executors; the HTEX multi-process executor may not have default capabilities to find and raise worker exceptions for the coordinating/initial Python process.

Solutions here might involve the following:

• Find and raise the root cause (or perhaps "lowest") exception when problems arise during a CytoTable run.
• Provide an avenue for adding this root cause information to a log file which is separate from or somehow easier to find / summarize than the Parsl logs.
• Consider adjusting the Parsl logs default behavior through CytoTable to avoid their output and focus on the summarized / more customized logging mentioned above.