Various fixes to CLEM workflow and bootstrap API endpoints

[tieneupin]

One thing led to another, and now there are a couple of fixes present in this PR:

• Loaded 'feedback_queue' from either the security config file or the TransportManager object. Resolves issue Some functions still look for 'feedback_queue' in MachineConfig #487 .
• Added unit tests for some of the functions fixed as part of this PR.
• Fixed some race conditions uncovered when refreshing database entries for the CLEM workflow.
• More verbose logging for CLEM workflow.
• Corrected canonical representation of variables in log messages for "bootstrap" API endpoints

Checks:

• CLEM workflow works on test setup

[codecov]

Codecov Report

Attention: Patch coverage is 6.77966% with 55 lines in your changes missing coverage. Please review.

Project coverage is 28.61%. Comparing base (400765c) to head (5acdbff).
Report is 1 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #494      +/-   ##
==========================================
+ Coverage   27.93%   28.61%   +0.67%     
==========================================
  Files          78       78              
  Lines       10255    10218      -37     
  Branches     1355     1354       -1     
==========================================
+ Hits         2865     2924      +59     
+ Misses       7295     7194     -101     
- Partials       95      100       +5     

[stephen-riggs]

Mostly looks fine. I'm a little unsure on the bits avoiding race conditions though. It would be nicer if you didn't have to do the waits, so is there a way to avoid doing the checks and refreshes?
I've noted the places you do the waits. If they're unavoidable then just dismiss the comments, but it feels like some of the checks aren't needed.

[tieneupin]

Mostly looks fine. I'm a little unsure on the bits avoiding race conditions though. It would be nicer if you didn't have to do the waits, so is there a way to avoid doing the checks and refreshes? I've noted the places you do the waits. If they're unavoidable then just dismiss the comments, but it feels like some of the checks aren't needed.

Your suggestion worked! I disabled all the db.refresh() entries and ran the workflow again, and it seems to be progressing with no failures. Looks like the refreshes were indeed unnecessary.

Your suggestion to remove the is_file() check from the validate_and_sanitise() functions was also sound. It caused more problems than it solved, since the file might not yet exist on the server-side at the time that check is called.

[stephen-riggs]

Thanks for making the changes, it looks much cleaner without the refreshes and sleeps. Looks good to me now :)