(Closes #3406) Fix subroutine handler to find the subroutine stmt if it exists instead of assuming its the first child.

[LonelyCat124]

@schreiberx This worked for the reproducer you gave us - if you have time to try it on NEMO itself it'd be appreciated.

Otherwise ready for a look from @arporter or @sergisiso. I will set ITs running in the mean time.

[LonelyCat124]

This PR also ticks fparser from 0.2.2 to >= 0.2.2 (since fparser has a 0.2.3 version now at least it helps us to use newer versions during development).

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 99.96%. Comparing base (1706d28) to head (6a8fad8).
⚠️ Report is 6 commits behind head on master.

Additional details and impacted files

@@           Coverage Diff           @@
##           master    #3407   +/-   ##
=======================================
  Coverage   99.96%   99.96%           
=======================================
  Files         389      389           
  Lines       54524    54542   +18     
=======================================
+ Hits        54505    54523   +18     
  Misses         19       19           

☔ 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.

[sergisiso]

@LonelyCat124 I have some doubts about the implementation and its test, see inline.

[LonelyCat124]

@sergisiso Ready for another look, I fixed the main changes not being tested properly but the fparser chjange I didn't revert yet as I'm confused as to why my local fparser reports a later version.

[sergisiso]

@LonelyCat124 See replies above, I like the new solution, there are just some minor comments now.

[LonelyCat124]

@sergisiso Reverted the setup.py change and added the additional checks into the tests.

[sergisiso]

@LonelyCat124 Everything looks good now and the integration test succeeded, this is approved for merging