[Bug] instantaneous_rate outputs a signal of one point less length

[dizcza]

Originally found in #362

import matplotlib.pyplot as plt
import numpy as np
import quantities as pq

from elephant import statistics, kernels
from elephant.spike_train_generation import homogeneous_poisson_process

np.random.seed(0)
spiketrain = homogeneous_poisson_process(rate=10 * pq.Hz,
                                         t_stop=10 * pq.s)
kernel = kernels.GaussianKernel(sigma=300 * pq.ms)
sampling_period = 445 * pq.ms
rate = statistics.instantaneous_rate(
    spiketrain,
    sampling_period=sampling_period,
    kernel=kernel, center_kernel=True, trim=False)

plt.plot(rate.times.simplified.magnitude, rate.magnitude, color='orange',
         label='instantaneous rate')
plt.eventplot(
    [spiketrain.simplified.magnitude, rate.times.simplified.magnitude],
    colors=['black', 'green'], lineoffsets=2, linelengths=1)
plt.legend()
plt.title(f"{kernel.__class__.__name__}: sigma={int(kernel.sigma.item())} ms, "
          f"sampling_period={int(sampling_period.item())} ms")
plt.show()

On the plot above, the green stripes denote rate.times, which lacks one point before the t_stop.