Pass no decoder cache to sub-models

The decoder cache does not work for host models due to having
no context manager, which is normally constructed by the
top-level network builder.

This commit fixes #207 as it does not globally disable the
decoder cache. We also add some asserts to existing tests
for ensure that #207 is fixed.

nengo_loihi/builder/builder.py

@@ -109,9 +109,12 @@ def __init__(self, dt=0.001, label=None, builder=None):
 
         # Host models filled in by the build process
         def create_host_model(label, dt):
+            # We don't use a decoder cache because it requires a context
+            # manager that differs depending on which sub-model is being built
             return NengoModel(
                 dt=float(dt),
                 label="%s, dt=%f" % (label, dt),
+                decoder_cache=NoDecoderCache(),
             )
 
         # TODO: these models may not look/behave exactly the same as

nengo_loihi/simulator.py

@@ -314,7 +314,6 @@ def __init__(  # noqa: C901
             raise ValidationError("network parameter must not be None",
                                   attr="network")
 
-        nengo.rc.set("decoder_cache", "enabled", "False")
         config.add_params(network)
 
         # ensure seeds are identical to nengo

nengo_loihi/tests/test_simulator.py

@@ -26,10 +26,15 @@ def test_model_validate_notempty(Simulator):
         a = nengo.Ensemble(10, 1)
         model.config[a].on_chip = False
 
+    assert nengo.rc.get("decoder_cache", "enabled")
+
     with pytest.raises(BuildError, match="No neurons marked"):
         with Simulator(model):
             pass
 
+    # Ensure cache config not changed
+    assert nengo.rc.get("decoder_cache", "enabled")
+
 
 @pytest.mark.parametrize("precompute", [True, False])
 def test_probedict_fallbacks(precompute, Simulator):