test: `PlannedGivingIntentScorer.predict_intent_score` never exercises its single-class fallback

[shivamlalakiya]

What's wrong

PlannedGivingIntentScorer.predict_intent_score short-circuits to
np.zeros(proba.shape[0]) when predict_proba returns fewer than two columns
(the single-class-seen-at-fit case). This is the same shape of guard as the one
fixed for PropensityScorer.predict_proba in #50, but in a different class and
method — nothing in tests/test_planned_giving.py fits on single-label data, so
this branch has never run.

Where

File	Locate with
philanthropy/models/_planned_giving.py	grep -n "proba.shape\[1\] < 2" philanthropy/models/_planned_giving.py

What to change

No source change — this is test-only.

1. Open tests/test_planned_giving.py and look at the existing
   predict_intent_score test (locate with
   grep -n "def test.*intent_score" tests/test_planned_giving.py) for fixture
   style.
2. Add a test that fits PlannedGivingIntentScorer on a y containing only one
   label, then calls predict_intent_score and asserts every score is 0.0.

Tests to add or extend

• File: tests/test_planned_giving.py
• Add: test_predict_intent_score_single_class_returns_zeros

Done when

python -m pytest tests/test_planned_giving.py --cov=philanthropy.models._planned_giving --cov-report=term-missing -q

philanthropy/models/_planned_giving.py reads 100% covered with an empty
Missing column.