Support residual evaluation from ceres::Problem

[B1ueber2y]

Follow up for #49. We can evaluate residuals either with EvaluateOptions or a list of residual ids.

Example:

import pyceres
import numpy as np

x1 = np.zeros(3)
x2 = np.ones(3)
x1_prior = np.arange(3)
covar = np.eye(3)
prob = pyceres.Problem()
loss = pyceres.TrivialLoss()
rid1 = prob.add_residual_block(pyceres.factors.NormalPrior(x1_prior, covar), loss, [x1])
rid2 = prob.add_residual_block(pyceres.factors.NormalError(covar), loss, [x1, x2])
print(prob.evaluate_residuals())
# Output:
# [0.0, -1.0, -2.0, -1.0, -1.0, -1.0]
print(prob.evaluate_residuals([rid2]))
# Output:
# [-1.0, -1.0, -1.0]