Propagate tags from fw_spec to metadata

[sivonxay]

Picking up #320.

I think the changed originally proposed would add alot of redundant information into the document/db.

[codecov]

Codecov Report

Merging #345 (75b92cd) into main (26608c5) will decrease coverage by 0.14%.
The diff coverage is 100.00%.

Additional details and impacted files

@@             Coverage Diff             @@
##              main     #345      +/-   ##
===========================================
- Coverage   100.00%   99.86%   -0.14%     
===========================================
  Files           19       19              
  Lines         1430     1439       +9     
  Branches       361      364       +3     
===========================================
+ Hits          1430     1437       +7     
- Misses           0        1       +1     
- Partials         0        1       +1     

Impacted Files	Coverage Δ
src/jobflow/managers/fireworks.py	100.00% <100.00%> (ø)

... and 1 file with indirect coverage changes

[sivonxay]

After speaking to @samblau, it seems that the functionality desired is to propagate the tags from the fw_spec into the metadata stored in the document.

While the documentation recommends to use Job.update_metadata({"tags": ["test"]}) function, people who are starting to use jobflow who have previously used fireworks+atomate are used to using the atomate.common.powerups.add_tags() function to add the tags. Because the tags aren't added to the metadata, they aren't propagated into the documents by jobflow.

[utf]

Thanks for this. I just had one minor comment.

[utf]

Maybe we should also check if tags already exist and if they are a list, and in that case append the tags?

[sivonxay]

Alright. Added.

One small issue I encountered here was that tags were getting duplicated if the job metadata had the "tags" key in it. Because of this line: spec.update(job.metadata) # add metadata to spec.

To account for this, I just removed all duplicates from job.metadata["tags"] assuming that duplicates will never be desired.

[utf]

That's a good catch. Thanks!