add timestep() predefined function akin to resolution()

doc/nestml_language/neurons_in_nestml.rst

@@ -266,7 +266,7 @@ The disadvantage of this method is that it requires a call to the ``resolution()
    equations:
        refr_t' = -1 / s    # a timer counting back down to zero
 
-Typically, the membrane potential should remain clamped to the reset or leak potential during the refractory period. It depends on the intended behavior of the model whether the synaptic currents and conductances also continue to be integrated or whether they are reset, and whether incoming spikes during the refractory period are taking into account or ignored.
+Typically, the membrane potential should remain clamped to the reset or leak potential during the refractory period. It depends on the intended behavior of the model whether the synaptic currents and conductances also continue to be integrated or whether they are reset, and whether incoming spikes during the refractory period are taken into account or ignored.
 
 In order to hold the membrane potential at the reset voltage during refractoriness, it can be simply excluded from the integration call:
 

tests/nest_tests/stdp_triplet_synapse_test.py

@@ -456,19 +456,7 @@ def _test_stdp_triplet_synapse(delay, spike_times_len):
     compare_results(ref_timevec, ref_w, nestml_timevec, nestml_w)
 
 
+@pytest.mark.parametrize("delay", [1., 5., 10.])
 @pytest.mark.parametrize("spike_times_len", [10])
-def test_stdp_triplet_synapse_delay_1(spike_times_len):
-    delay = 1.
-    _test_stdp_triplet_synapse(delay, spike_times_len)
-
-
-@pytest.mark.parametrize("spike_times_len", [10])
-def test_stdp_triplet_synapse_delay_5(spike_times_len):
-    delay = 5.
-    _test_stdp_triplet_synapse(delay, spike_times_len)
-
-
-@pytest.mark.parametrize("spike_times_len", [10])
-def test_stdp_triplet_synapse_delay_10(spike_times_len):
-    delay = 10.
+def test_stdp_triplet_synapse_delay_1(spike_times_len, delay):
     _test_stdp_triplet_synapse(delay, spike_times_len)