fix: fixes issues #225

[baszalmstra]

Fixes an issue where taking the field of a temporary crashed the compiler. This was caused by a field expression always expecting a "place" expression which is actually not always the case (in the case of a temporary for example).

Closes #225

[codecov]

Codecov Report

Merging #232 into master will increase coverage by 0.00%.
The diff coverage is 86.20%.

@@           Coverage Diff           @@
##           master     #232   +/-   ##
=======================================
  Coverage   81.13%   81.14%           
=======================================
  Files         182      182           
  Lines       11938    11968   +30     
=======================================
+ Hits         9686     9711   +25     
- Misses       2252     2257    +5     

Impacted Files	Coverage Δ
crates/mun_codegen/src/ir/body.rs	84.87% <85.96%> (-0.10%)	⬇️
crates/mun_codegen/src/test.rs	98.24% <100.00%> (+0.01%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update b738800...8078734. Read the comment docs.

[Wodann]

Future Work
This is a great example of where we can optimise a lot by using a struct(value) instead of a struct(gc). For temporaries (or scoped struct instances) it doesn't really make sense to allocate them on the heap.

[baszalmstra]

That would be great! But the use case is quite small, it's only when you construct a GC struct that doesn't leave the scope of the function at all (otherwise you can't make guarantees). Nonethess, if we can catch that in the compiler and turn it into a stack allocation, that would be great!