Zeiss CZI: do not set indexed for RGB images

[melissalinkert]

See QA 10987, fixes gh-1719.

To test, verify that opening the file from QA 10987 in Matlab (as described in the issue) or via showinf -separate no longer throws an exception. The image is quite large, so you may need to increase the heap size or crop the image to prevent an OOM.

[sbesson]

Tested directly from the branch (note this PR will need to go through the merge builds tonight). The ChannelSeparator reader is correctly initialized on the file from QA 10987 and all planes are read without any exception thrown for all series (tested using bfopen).
Data quality is inconsistent between series e.g. series 9 renders fine but series 0 looks buggy. Not sure whether this arises from the original file and/or is covered by other CZI bug fixes.

[melissalinkert]

This should probably be retested in conjunction with gh-1725 or as part of the merge build - with gh-1725 included, there should only be 1 series that represents the full image, and I'd expect it to more or less look like it does in Zen.

[mtbc]

It's much faster to showinf this image from a local filesystem than over CIFS! Things seem good with the merge build. Opening had,

Initializing reader
ZeissCZIReader initializing /tmp/IP-Stitching-01-Change Scaling-01.czi
Unknown IlluminationType value 'Fluorescence' will be stored as "Other"
Initialization took 0.335s

Reading core metadata
Filename = /tmp/IP-Stitching-01-Change Scaling-01.czi
Series count = 1
Series #0 :
    Image count = 3
    RGB = false (1) (separated)
    Interleaved = true
    Indexed = false (false color)
    Width = 12393
    Height = 9369
    SizeZ = 1
    SizeT = 1
    SizeC = 3
    Thumbnail size = 128 x 96
    Endianness = intel (little)
    Dimension order = XYCZT (certain)
    Pixel type = uint16
    Valid bits per pixel = 14
    Metadata complete = false
    Thumbnail series = false
    -----
    Plane #0 <=> Z 0, C 0, T 0
    Plane #1 <=> Z 0, C 1, T 0
    Plane #2 <=> Z 0, C 2, T 0


Reading pixel data (0-2)
    Read 1/3 planes (33%)
    Read 2/3 planes (66%)
    Read 3/3 planes (100%)
[done]
217.266s elapsed (72422.0ms per plane)

Launching image viewer

only one series and no exceptions. Other CZI files still seem fine.

Trying it with -thumbs doesn't look too good, perhaps that's expected?
With -separate:

Without -separate:

[sbesson]

Note pyramid completion time for the file above was:

2015-04-16 18:25:43,555 INFO  [ ome.services.pixeldata.PixelDataHandler] (2-thread-1) HANDLED EventLog:3907867(entityId=28314) [24851493 ms.]

[mtbc]

The resulting image looks good, though. The thumbnail from Insight is:

[melissalinkert]

Any objections to merging this? The -thumbs behavior is outside the scope of this PR.

[mtbc]

No objections, go for it! 👍

[bramalingam]

Tested with Matlab and bftools (showinf -nopix) and in both cases there were no exceptions. Good to merge..

[melissalinkert]

--no-rebase