feat: add Prometheus metrics for Redis cache

Adding a Prometheus counter to be able to monitor cache hits/misses
for search queries

Signed-off-by: moson-mo <mo-son@mailbox.org>

test/test_metrics.py

@@ -0,0 +1,40 @@
+import pytest
+from prometheus_client import REGISTRY, generate_latest
+
+from aurweb import db
+from aurweb.cache import db_query_cache
+from aurweb.models.account_type import USER_ID
+from aurweb.models.user import User
+
+
+@pytest.fixture(autouse=True)
+def setup(db_test):
+    return
+
+
+@pytest.fixture
+def user() -> User:
+    with db.begin():
+        user = db.create(
+            User,
+            Username="test",
+            Email="test@example.org",
+            RealName="Test User",
+            Passwd="testPassword",
+            AccountTypeID=USER_ID,
+        )
+    yield user
+
+
+@pytest.mark.asyncio
+async def test_search_cache_metrics(user: User):
+    # Fire off 3 identical queries for caching
+    for _ in range(3):
+        await db_query_cache("key", db.query(User))
+
+    # Get metrics
+    metrics = str(generate_latest(REGISTRY))
+
+    # We should have 1 miss and 2 hits
+    assert 'search_requests_total{cache="miss"} 1.0' in metrics
+    assert 'search_requests_total{cache="hit"} 2.0' in metrics