fix(opentargets): adapt to upstream API drift + skip deprecated expressions field

[lauraluebbert]

Fixes the 11 opentargets tests failing across all Python versions on main. Three distinct upstream changes — addressed independently:

1. GraphQL schema change (real code bug) — 2 tests

RuntimeError: OpenTargets GraphQL returned HTTP 400.
Body: Field 'synonyms' of type '[DrugLabelAndSource!]!' must have a sub selection.

The Drug type's `synonyms` and `tradeNames` are now `[DrugLabelAndSource!]!` instead of scalar lists. Querying without a sub-selection 400s. Updated `QUERY_STRING_DRUGS` to query `{ label }` — the existing `_collapse_singletons()` post-processor flattens each `{label: "X"}` back to `"X"` automatically, so the public DataFrame shape is preserved (`drug.synonyms` remains `list[str]`).

Confirmed via GraphQL introspection: `DrugLabelAndSource.{label, source}` are both `String!`.

2. Data drift — 8 fixtures refreshed

Per project convention (refresh fixtures, don't normalize):

Test	What changed upstream
`test_opentargets`	Disease ontology: `EFO_0000274` → `MONDO_0004980`
`test_opentargets_diseases`	Same disease ID change
`test_opentargets_drugs`	Re-captured with the synonyms fix applied
`test_opentargets_interactions`	Protein IDs (e.g. `ENSP00000361004` → `ENSP00000360730`)
`test_opentargets_pharmacogenetics`	Variant allele frequencies (`T_C,C` → `T_C,T`)
`test_opentargets_depmap`	Hash refreshed (1239 rows)
`test_opentargets_depmap_filter`	UBERON_0002367 (prostate) has no depmap data for IL13 anymore; fixture now expects `[]`
`test_opentargets_interactions_no_limit`	Hash refreshed (25 rows)

Refreshed via a one-shot script that calls each function with the test args and rewrites `expected_result` to match.

3. `expressions` field deprecated — 2 tests skipped

Open Targets removed data from `Target.expressions` (returns `[]` for all queries). The replacement is `Target.baselineExpression` with a completely different schema:

• Old: `expressions[].tissue.{id, label, anatomicalSystems, organs}` + `expressions[].rna.{zscore, value, unit, level}`
• New: `baselineExpression.rows[].{targetFromSourceId, max, q1, q3, median, min, tissueBiosample, celltypeBiosample, datasourceId, …}`

Migrating `gget_opentargets` to the new field is a user-facing API change (different DataFrame columns) that warrants its own PR. For now:

• `test_opentargets_expression` and `test_opentargets_expression_no_limit` are marked `type: "skip"` with a `reason` field
• `tests/from_json.py` gains a new `skip` test type so JSON-defined tests can be skipped with a reason message (no need for a separate Python test file)

[codecov-commenter]

Codecov Report

❌ Patch coverage is 0% with 1 line in your changes missing coverage. Please review.
✅ Project coverage is 7.98%. Comparing base (682d7c6) to head (daeee55).
⚠️ Report is 5 commits behind head on dev.

Files with missing lines	Patch %	Lines
gget/gget_opentargets.py	0.00%	1 Missing ⚠️

Additional details and impacted files

@@           Coverage Diff            @@
##             dev    #256      +/-   ##
========================================
- Coverage   7.98%   7.98%   -0.01%     
========================================
  Files         29      29              
  Lines       9244    9245       +1     
========================================
  Hits         738     738              
- Misses      8506    8507       +1     

☔ 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.