Add support for saving string stash codes to PP

[rcomer]

🚀 Pull Request

Description

Closes #6239. String stash codes now "just work".

---

Consult Iris pull request check list

---

Add any of the below labels to trigger actions on this PR:

• benchmark_this Request that this pull request be benchmarked to check if it introduces performance shifts

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 89.85%. Comparing base (331f012) to head (bc48dc1).
Report is 71 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #6289      +/-   ##
==========================================
- Coverage   89.85%   89.85%   -0.01%     
==========================================
  Files          88       88              
  Lines       23384    23387       +3     
  Branches     4356     4357       +1     
==========================================
+ Hits        21012    21014       +2     
  Misses       1646     1646              
- Partials      726      727       +1     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[rcomer]

Ah linkcheck, I have not missed you.

[pp-mo]

Ah linkcheck, I have not missed you.

😞 I think we should remove this from PRs into a periodic check like benchmarks
#6292

[pp-mo]

I think this is totally good enough.

I did wonder about replacing arbitrary odd object with string reprs in netcdf save, but I think that is a step too far : STASH is a special recognised attribute name, and we also convert them back on loading, so we can save UM data and re-save correctly as PP fields with the original stashes.
FWIW we should probably so the same for GRIB too, I currently have an issue open about that.