Minor xspec additions

[DougBurke]

Summary

Minor changes to the XSPEC module, including added documentation to the set_xsabund routine to note the availability of the "felc" table, and set_xsstate no-longer requires all fields to be present.

Details

This is a small subset of the changes from #1615 that do not require changing the C++ interface. The changes are (not in order, and not per commit):

• an internal change to the tests so that mypy is less picky
• the read_xstable_model routine can now be sent a filename as a Path object as well as a string; this is an "internal" routine so it is not going to change most people's lives
• set_xsabund documentation is updated to mention the "felc" table
• get_xsabund has an extra example in its docstring
• updated the typing in sherpa/astro/xspec/__init.py to use Python 10 syntax (can avoid Union and Optional)
  • as part of this, use an overload for get_xsabund
  • which means should we annotate the non-overload version; previously we have not done this because why should we, but
    • I am reading proposed new documentation Update to the Overloads chapter python/typing#1839 which doesn't exactly spell it out
    • however, by annotating all versions we do guarantee that help(set_xsabund) will see the types
    • it also stops a pyright complaint
    • and seems to mesh with this discussion (that it makes sense to add the types): https://discuss.python.org/t/conflict-between-python-typing-library-docs-and-mypy-docs-regarding-function-overload-usage/75087
• set_xsstate used to require all settings are present before it would run, which can be useful, but it also means that if we add a new field (as I plan to) , or delete an old field, then old states can not be restored because it wouldn't know about the new settings. So we now allow each setting to be optional.

[codecov]

Codecov Report

Attention: Patch coverage is 85.10638% with 7 lines in your changes missing coverage. Please review.

Project coverage is 87.47%. Comparing base (c00eeff) to head (f002119).
Report is 18 commits behind head on main.

Files with missing lines	Patch %	Lines
sherpa/astro/xspec/__init__.py	85.10%	7 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #2215      +/-   ##
==========================================
+ Coverage   87.44%   87.47%   +0.02%     
==========================================
  Files          84       84              
  Lines       29848    29860      +12     
  Branches     3833     3831       -2     
==========================================
+ Hits        26101    26119      +18     
+ Misses       3637     3631       -6     
  Partials      110      110              

Files with missing lines	Coverage Δ
sherpa/astro/xspec/__init__.py	98.25% <85.10%> (-0.07%)	⬇️

... and 2 files with indirect coverage changes

---

Continue to review full report in Codecov by Sentry.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update c00eeff...f002119. Read the comment docs.

[DougBurke]

@hamogu - I made a small change to a test since you accepted it (found when working on #2216 but worth cleaning up here). Can you just give the last commit a quick check? Ta

[hamogu]

I don't see any functional change, only a renaming from key to miss_key. I mean, who not? But what's the intention?

[DougBurke]

Because it separates the input argument (that is parametrized, aka miss_key) from the generic variable name key. So I can use key in the rest of the loop and not get strange behaviour (as ended up happening in later work where the code would fail because I'd not separated the two concepts).

[hamogu]

Makes sense - and that subtlety also explains why I didn't notice the change on review, when Gh highlights just the changes lines and not the others where the word "key" also appears.