editoast: return conflicts list when STDCM request fails

[hamz2a]

closes #8639

[codecov-commenter]

⚠️ Please install the to ensure uploads and comments are reliably processed by Codecov.

Codecov Report

Attention: Patch coverage is 87.27915% with 36 lines in your changes missing coverage. Please review.

Project coverage is 39.52%. Comparing base (5977531) to head (a19b328).
Report is 8 commits behind head on dev.

Files with missing lines	Patch %	Lines
editoast/src/views/timetable/stdcm.rs	86.95%	36 Missing ⚠️

❗ Your organization needs to install the Codecov GitHub app to enable full functionality.

Additional details and impacted files

@@             Coverage Diff              @@
##                dev    #9327      +/-   ##
============================================
+ Coverage     39.07%   39.52%   +0.45%     
  Complexity     2270     2270              
============================================
  Files          1308     1308              
  Lines         99317    99588     +271     
  Branches       3316     3316              
============================================
+ Hits          38805    39363     +558     
+ Misses        58547    58260     -287     
  Partials       1965     1965              

Flag	Coverage Δ
core	74.99% <ø> (ø)
editoast	73.24% <87.23%> (+1.32%)	⬆️
front	10.35% <ø> (+0.03%)	⬆️
gateway	2.50% <ø> (ø)
osrdyne	3.29% <ø> (ø)
railjson_generator	87.49% <ø> (ø)
tests	86.71% <100.00%> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

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

[woshilapin]

Again some more debuggability information send back to the front, I love it 😍

In this PR, there is some renaming interlaced with changes of functionality. It would have been nice to first do the renaming in an independent commit to ease the review of the important part (note that the renaming is nice, just not a change that need attentive reviewing).

[leovalais]

Thanks for this PR! Left a few comments.

On a more general note, the /stdcm endpoint is starting to do a whole lot of things and this process isn't documented anywhere. The endpoint documentation is just

/// Compute a STDCM and return the simulation result

which isn't exact anymore following your changes.

I'd suggest:

• documenting the whole process in the handler's documentation with the "happy path" and what we're doing if the simulation fails, making sure to separate both
• continue the series of scoping comments // n. Nth step in your additions, as it's really helpful to figure out where we're at in the global process
• split the individual steps into documented functions (you don't have to do that for existing steps, but it would be a nice thing to do for your changes so that we can start enforcing a better modularity from now on)

Wdyt?

[shenriotpro]

LGTM for tests/

[woshilapin]

Sorry for this second pass of review, but I feel we lack some improvements of code architecture. Let's see together if we can make the code more simple and maintainable.

[leovalais]

LGTM, the remaining comments will be addressed in a separate PR with a refactoring of the endpoint and its utilities.

[woshilapin]

Let's merge this one and address all the unaddressed comments in the follow-up PR.