Add mock catalog type requiring no calibration, and add example pipeline

[joezuntz]

Addresses issue #360

Still looking at the ingestion part.

[ts485]

Hi Joe,

We also need an output "source_photoz_pdfs" (which is originally generated from the stage "BPZ_lite") to run the stage "CLClusterShearCatlog". Can either create one more stage or add the functionality to "TXSourceSelectorSimple" to generate "source_photoz_pdfs"?

I attached the flowchart of CL pipeline and the mock pipeline.

Best,
Tomomi

[joezuntz]

I will add the BPZ stages to the pipeline, I think it should work fine.

[joezuntz]

For speed, and to avoid difficult questions about photo-z for the source sample (which may have only r, i ,z), I have added a quick stage to generate fake source_photoz_pdfs using the true redshifts. I've added it to the pipeline file.

I've also added a call to the CLClusterShearCatalogs stage. To use that as-is you will need to download and extract the latest version of the example.tar.gz inputs file. See the README for the URL.

[ts485]

I ran the pipeline and it generated the files properly. I have not checked output files yet, but it seems working fine.

[ts485]

@joezuntz Hi Joe, I realized that now I can use the latest version of CLMM in TXPipe at NERSC, but now I cannot run the stage "CLClusterShearCatalogs". Have you checked whether the pipeline works fine?

[ts485]

P.S. I could run the pipeline till the end by using tx-10 instead of tx