update-traits crash when traits malformed

[sentry]

Sentry Issue: FLAGSMITH-API-68J

TypeError: util.engine_models.identities.traits.models.TraitModel() argument after ** must be a mapping, not list
(4 additional frame(s) were not displayed)
...
  File "edge_api/identities/views.py", line 180, in update_traits
    trait = TraitModel(**request.data)

Acceptance criteria

• update-traits data validation broadens and responds 400 when parameters are malformed, e.g. traits is given a JSON array instead of an object.

[m-husnain-dev]

Hi, I'd like to work on this issue. I'll add validation to handle the case where request.data isn't a mapping, and return a proper 400 error instead of crashing.

[zain-asif-dev]

Opened a fix here: #7952

The root cause is that TraitModel(**request.data) raises a bare TypeError (not pydantic.ValidationError) when request.data is a list rather than a dict, so it slips past the existing error handling and becomes a 500. Fixed by validating the payload shape before it reaches pydantic and switching to TraitModel.model_validate(), which also uncovered a related crash in pyngo.drf_error_details() for that same case. Added a regression test that reproduces the original 500 and confirms it now returns a 400.