LonelyCat124

Ready for a first review.

@sergisiso should be ready for another look now I think - one comment I didn't address as I'm unclear what you meant - I tried to remove `@property` but then...

N.B. #2039 and #2040 were known issues with the previous implementation, we need to be smarter with control flow for this implementation (easy option is to give up if we...

> But tbh, I don't fully understand what the "write" alone does here, presumably we have to fence the operation from the read to the write, otherwise we don't solve...

re: collapse + compilers maybe we should test with: Recent-ish gcc (9 or 10?) Recent-ish intel (2021? 2022?) Recent-ish Nvidia Recent-ish cray (if we have access somewhere?) MO current preferred...

I think this is a duplicate of issues mentioned in #1841 (second comment)

@hiker Quick question on how the ordering of statements goes with the dependency system - if you have an assignment such as `a = a + 1` does the `a`...

@sergisiso this is ready for a first look. I didn't add anything to user or dev guide for now, let me know if you think I should. I'm waiting for...

@sergisiso I've addressed most of the comments here - I think the next_access one is a bigger issue than just here (and we need to decide if the variable access...

A simple standalone fail case again: ``` def test_fail_case(fortran_reader): code = ''' real function my_func(a1, a2) real a1, a2 my_func = a1 * a2 end function subroutine a(b, c, d)...