D_eta: Create a design matrix D for a model eta
In yingboli/BayesMDL: Bayesian MDL for Multiple Changepoint Detection

Create a design matrix D for a model eta

1	D_eta(x, eta, scale_trend_design)

`x`	The time series data, a numeric vector of length `n`.
`eta`	A multiple changepoint configuration, i.e., model. It is a 0/1 indicator vector of length `n`, with the first `p` elemenets always being 0.
`scale_trend_design`	The factor multiplied to the design matrix of trend. Default is 1/50.

`D`	A `n` by `(2*m-1)` matrix. The first `m` columns are for regime means; they are 0/1 indicators for regimes from 2 to `m`. The next `(m+1)` columns are for trends.
`seg_lengths`	A vector of length `m`, regime lengths, including the first regime.