fix: Take all spans into account when determining snippet alignment

[mark-koch]

No description provided.

[codecov-commenter]

Codecov Report

Attention: Patch coverage is 50.00000% with 3 lines in your changes missing coverage. Please review.

Project coverage is 91.70%. Comparing base (8134a85) to head (d760742).
Report is 1 commits behind head on feat/diagnostics.

Files with missing lines	Patch %	Lines
guppylang/definition/function.py	0.00%	3 Missing ⚠️

Additional details and impacted files

@@                 Coverage Diff                  @@
##           feat/diagnostics     #588      +/-   ##
====================================================
- Coverage             91.73%   91.70%   -0.03%     
====================================================
  Files                    61       61              
  Lines                  6509     6513       +4     
====================================================
+ Hits                   5971     5973       +2     
- Misses                  538      540       +2     

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

[acl-cqc]

Looks good to me - basically, I am just feeling silly that I was confused. But a couple of suggestions if you like :)

[acl-cqc]

nit: took me awhile to figure out what this * was about - it does seem like there's one more construction/destruction than I'd expect. It might be less efficient but [span] + [to_span(child.span) for child in diag.children if child.span] might be clearer. Up to you!

[acl-cqc]

This confused me a bit too - I realize this is purely formatting, but this looks like the subdiagnostic should also refer to line 99! (What happens if it does? Is that a separate test worth having?)

Changing the "This is an apple" to "This apple is on another line" might have avoided confusion ;)