Possible typo in float-16 NaN encoding #53

In current master/src/cbor/encoding.c line 140, there appears to be a constant encoded value for 16-bit float NaN value as 0x00e700. I think this is a typo as RFC 7049 in Section 3.9 states that

If NaN is an allowed value, it must always be represented as 0xf97e00

Notice the first data octet difference e7 vs 7e.
The same 7e00 encoding comes up in the RFC 7049 NaN example in Appendix A.