SIDD 2.0 Compression Schema Reading Bug

[ktopolov]

It seems there might be a bug in XML metadata reading for SIDD2.0+ for the "Compression" field. It seems from the specification that this is the expected formatting:

    <Compression>
        <J2K>
            <Original>
                <NumWaveletLevels>6</NumWaveletLevels>
                <NumBands>1</NumBands>
                <LayerInfo numLayers="2">
                    <Layer index="1">
                        <Bitrate>2.00000000000000000E01</Bitrate>
                    </Layer>
                    <Layer index="2">
                        <Bitrate>2.00000000000000000E01</Bitrate>
                    </Layer>
                </LayerInfo>
            </Original>
        </J2K>
    </Compression>

and C++ Six library writes it this way, but sarpy fails to read with the error:

Attribute Custom of class FilterKernelType received <Element '{urn:SIDD:2.0.0}Custom' at 0x7f101f671b20>,
	but values ARE REQUIRED to be one of ('GENERAL', 'FILTER BANK')
Failed converting <Element '{urn:SIDD:2.0.0}Original' at 0x7f101f6997b0> of type <class 'xml.etree.ElementTree.Element'> to Serializable type <class 'sarpy.io.product.sidd2_elements.Compression.J2KSubtype'>
	for field Original of class J2KType with exception <class 'KeyError'> - 'size'.
	Setting value to None, which may be against the standard.

It seems that sarpy is instead expecting numLayers to instead be size, or it may use the wrong "descriptor" type altogether - it appears to simply want an array of bitrates rather than these individual indexed layers.

[ARKAMarkDologuin]

I've added a parser that goes through to the different Layers and gets the bitrates to place into an array as per LayerInfo object expectations
ARKAMarkDologuin@e5ab5e3

[petersontex]

Please test this with the new v1.3.62 release candidate.

The code is here: https://github.com/ngageoint/sarpy/releases/tag/v1.3.62rc1

Also available from PyPi (As a full release) here: https://pypi.org/project/sarpy/