Compare effective schemas in oasdiff (flatten-allof)

[turegjorup]

Enables flatten-allof: true on both oasdiff steps in the API-spec workflow.

Changes

• .github/workflows/api-spec.yml: add flatten-allof: true to the breaking and changelog steps.

Why — and an honest note on scope

flatten-allof compares the effective (flattened) schema, so an allOf/$ref restructure isn't misreported as removed properties. This is the class of noise the 4.1→4.3 upgrade produced (7× "removed required hydra:member" etc. from the new HydraCollectionBaseSchema).

I verified empirically against the real 4.1→4.3 spec diff:

	errors	warnings
without flatten	70	84
with flatten	113	0

So it does eliminate the ~98 allOf artifacts — but it does not reduce the gate to near-zero: the 4.1→4.3 error-schema widenings (nullable/type changes) remain and even expand once the effective schemas are compared field-by-field. Those are real, and that entire diff is a merged one-off — going forward every comparison is 4.3-vs-4.3, so the allOf-mismatch class can't recur regardless of this flag.

Net: this is forward-looking hygiene (robustness against future API Platform schema restructures), not a fix for a present problem. It's low-risk — flatten-allof never masks a real breaking change; it only changes how structurally-refactored-but-equivalent schemas are compared.

[codecov-commenter]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 72.99%. Comparing base (191e1ed) to head (977a675).
⚠️ Report is 15 commits behind head on develop.

Additional details and impacted files

@@            Coverage Diff             @@
##             develop      #53   +/-   ##
==========================================
  Coverage      72.99%   72.99%           
  Complexity       232      232           
==========================================
  Files             29       29           
  Lines            648      648           
==========================================
  Hits             473      473           
  Misses           175      175           

Flag	Coverage Δ
unittests	72.99% <ø> (ø)

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

☔ View full report in Codecov by Harness.
📢 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.