fix(api): filter by every GIAS phase, not just the grouped ones

/api/schools only recognised primary, secondary and all-through. Any
other phase the search page offers (nursery, 16 plus, middle deemed
primary/secondary) fell through to no filter, so "Nursery" returned the
whole result set, mostly primaries. Ungrouped phases now match exactly,
and an unknown phase returns nothing rather than everything.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

backend/app.py

@@ -736,7 +736,7 @@ async def get_schools(
         None, description="Filter by local authority", max_length=100
     ),
     school_type: Optional[str] = Query(None, description="Filter by school type", max_length=100),
-    phase: Optional[str] = Query(None, description="Filter by phase: primary, secondary, all-through", max_length=50),
+    phase: Optional[str] = Query(None, description="Filter by phase: primary or secondary (grouped), or any GIAS phase name (exact)", max_length=50),
     postcode: Optional[str] = Query(None, description="Search near postcode", max_length=10),
     radius: float = Query(5.0, ge=0.1, le=5, description="Search radius in miles"),
     page: int = Query(1, ge=1, le=1000, description="Page number"),
@@ -771,11 +771,13 @@ async def get_schools(
 
     # Phase filter — uses PHASE_GROUPS so all-through/middle schools appear
     # in the correct phase(s) rather than being invisible to both filters.
+    # Any other GIAS phase (nursery, 16 plus, middle deemed ...) is an exact
+    # match. It must never fall through to no filter: the search page offers
+    # every phase, and "Nursery" used to return the whole result set.
     if phase:
         phase_lower = phase.lower().replace("_", "-")
-        allowed = PHASE_GROUPS.get(phase_lower)
-        if allowed:
-            df_latest = df_latest[df_latest["phase"].str.lower().isin(allowed)]
+        allowed = PHASE_GROUPS.get(phase_lower, {phase_lower})
+        df_latest = df_latest[df_latest["phase"].fillna("").str.lower().isin(allowed)]
 
     # Secondary-specific filters (after phase filter)
     if gender:

backend/tests/test_phase_filter.py

@@ -0,0 +1,84 @@
+"""The /api/schools phase filter.
+
+The search page offers every GIAS phase, but the filter only knew the three
+grouped ones (primary, secondary, all-through). Anything else — nursery,
+16 plus, the middle-deemed phases — fell through to no filter at all, so
+"Nursery" returned the whole result set, mostly primaries.
+"""
+
+import numpy as np
+import pandas as pd
+import pytest
+from fastapi.testclient import TestClient
+
+PHASES = {
+    100001: "Nursery",
+    100002: "Primary",
+    100003: "Middle deemed primary",
+    100004: "Secondary",
+    100005: "Middle deemed secondary",
+    100006: "16 plus",
+    100007: "All-through",
+}
+
+
+def _schools_df() -> pd.DataFrame:
+    base = {
+        "local_authority": "Testshire",
+        "school_type": "Academy",
+        "address": "1 Test Street",
+        "town": "Testtown",
+        "postcode": "TS1 1AA",
+        "religious_denomination": None,
+        "age_range": "4-11",
+        "has_sixth_form": None,
+        "gender": "Mixed",
+        "admissions_policy": None,
+        "ofsted_grade": np.nan,
+        "ofsted_date": None,
+        "ofsted_framework": None,
+        "latitude": 51.5,
+        "longitude": -0.1,
+        "year": 202425,
+        "total_pupils": 300,
+        "rwm_expected_pct": np.nan,
+        "attainment_8_score": np.nan,
+    }
+    return pd.DataFrame([
+        {**base, "urn": urn, "school_name": f"{phase} School", "phase": phase}
+        for urn, phase in PHASES.items()
+    ])
+
+
+@pytest.fixture()
+def client(monkeypatch):
+    from backend import app as app_module
+
+    monkeypatch.setattr(app_module, "load_latest_school_data", _schools_df)
+    monkeypatch.setattr(app_module, "load_school_data", _schools_df)
+    return TestClient(app_module.app, raise_server_exceptions=False)
+
+
+def _urns(client, phase):
+    resp = client.get("/api/schools", params={"phase": phase})
+    assert resp.status_code == 200, resp.text
+    return sorted(s["urn"] for s in resp.json()["schools"])
+
+
+@pytest.mark.parametrize("phase, urn", [
+    ("nursery", 100001),
+    ("16 plus", 100006),
+    ("middle deemed primary", 100003),
+    ("middle deemed secondary", 100005),
+])
+def test_an_ungrouped_phase_matches_exactly(client, phase, urn):
+    assert _urns(client, phase) == [urn]
+
+
+def test_grouped_phases_still_take_in_their_related_phases(client):
+    assert _urns(client, "primary") == [100002, 100003, 100007]
+    assert _urns(client, "secondary") == [100004, 100005, 100006, 100007]
+
+
+def test_an_unknown_phase_returns_nothing_rather_than_everything(client):
+    assert _urns(client, "kindergarten") == []