Chebyshev polynomials don't handle special case: n=1

[Paul Price]

The special case n=1 for polynomials (i.e., simply a constant) is important to
be able to handle --- we often treat the order of the polynomial as something
that will be tuned to handle the data, and sometimes simply setting it to just a
constant is the best option.

Evaluating a chebyshev polynomial where any dimension has the order set to 1
produces a SEGV because the function createChebyshevPolys does not treat the
special case, but attempts to immediately create higher order terms:

Create the Chebyshev polynomials.
Polynomial i has i-th order.
chebPolys[0]->coeff[0] = 1;
chebPolys[1]->coeff[1] = 1;

The specific case encountered was for a psDPolynomial2D with nX = nY = 1 using
rel3, but inspection of the code reveals that it is still an issue for rel4.

Testing, performed in tst_psFunc00.c, covers only a single set of polynomial orders.

[gusciora@…]

This should be fixed in the current development version of the code. Tests must
still be augmented to handle order-1 polynomials.

[gusciora@…]

This should be done in the current development version of the code.

Must improve tests to handle order-1 polynomials.

[Paul Price]

Bug has been resolved.... closing.

[Paul Price]

Bugs have been fixed... closing.