fix(hwfit): tolerate non-numeric gpu_count in /api/hwfit/models (#3639)

* fix(hwfit): tolerate non-numeric gpu_count in /api/hwfit/models

The route did `n = int(gpu_count)` with no guard, so a non-numeric query param
like `?gpu_count=abc` raised ValueError and returned HTTP 500. Parse it
defensively (mirroring the gpu_group guard a few lines above): a malformed value
is ignored, exactly like omitting the param, and valid values still apply.

Adds tests/test_hwfit_gpu_count_nonnumeric.py: a non-numeric gpu_count returns a
ranking instead of raising, and a numeric value is still accepted.

* test(hwfit): cover non-numeric manual_gpu_count too

Follow-up to the gpu_count guard: add a regression test for the sibling
manual_gpu_count query param (the hardware simulator in _apply_manual_hardware),
which dev already guards by defaulting to 1 on a non-numeric value. This pins
that behaviour so the endpoint's count parsing is fully covered and cannot
regress to a 500.

routes/hwfit_routes.py

@@ -177,8 +177,14 @@ def setup_hwfit_routes():
             system["gpu_name"] = g["name"]
             system["active_group"] = {**g, "use_count": n}
 
-        if gpu_count != "":
+        # Parse the optional count defensively (matches the gpu_group guard
-            n = int(gpu_count)
+        # above): a non-numeric query param previously raised ValueError ->
+        # HTTP 500. A malformed value is ignored, same as omitting it.
+        try:
+            n = int(gpu_count) if gpu_count != "" else None
+        except ValueError:
+            n = None
+        if n is not None:
             if n == 0:
                 # RAM-only mode: rank against system memory, offload allowed.
                 system["has_gpu"] = False

tests/test_hwfit_gpu_count_nonnumeric.py

@@ -0,0 +1,38 @@
+"""GET /api/hwfit/models must not 500 on a non-numeric gpu_count.
+
+The handler did `n = int(gpu_count)` with no guard, so `?gpu_count=abc` (or any
+non-integer) raised ValueError -> HTTP 500. A malformed count is now ignored,
+matching how the neighbouring gpu_group param is already parsed.
+"""
+from routes.hwfit_routes import setup_hwfit_routes
+
+
+def _get_models():
+    router = setup_hwfit_routes()
+    for route in router.routes:
+        if getattr(route, "path", "").endswith("/models") and "GET" in getattr(route, "methods", set()):
+            return route.endpoint
+    raise AssertionError("hwfit /models route not found")
+
+
+def test_non_numeric_gpu_count_does_not_raise():
+    handler = _get_models()
+    # Previously raised ValueError (HTTP 500); now degrades to a normal ranking.
+    result = handler(gpu_count="abc")
+    assert isinstance(result, dict)
+
+
+def test_numeric_gpu_count_still_accepted():
+    handler = _get_models()
+    result = handler(gpu_count="0")
+    assert isinstance(result, dict)
+
+
+def test_non_numeric_manual_gpu_count_does_not_raise():
+    # manual_gpu_count is the other count param on this endpoint (the hardware
+    # simulator in _apply_manual_hardware). A non-numeric value must also degrade
+    # (default to 1) rather than 500, so the endpoint's count parsing is fully
+    # covered.
+    handler = _get_models()
+    result = handler(manual_mode="gpu", manual_gpu_count="abc")
+    assert isinstance(result, dict)