MRO - rework overall report creation (and single store) logic

[clemensutschig]

Today single repo docs (tir/dtr) are put into the json tree that represent the project.. this has two major drawbacks:

1. We cannot dump the tree into the overall docs where it would belong, as the bytes are still there and
2. we reinvent jenkins std functionality.. namely 'stash

So rather than putting byte[] into the Project json structure we should put in the names of the docs we stash and use that to generate the overall reports..

References:

1. Byte[] into Structure: https://github.com/opendevstack/ods-jenkins-shared-library/blob/master/src/org/ods/orchestration/usecase/LeVADocumentUseCase.groovy#L1143

2. Overall.. byte [] out of Structure: https://github.com/opendevstack/ods-jenkins-shared-library/blob/master/src/org/ods/orchestration/usecase/DocGenUseCase.groovy#L83

3. Finalize.. toString() contains the byte[] : https://github.com/opendevstack/ods-jenkins-shared-library/blob/master/src/org/ods/orchestration/FinalizeStage.groovy#L65

If we remove the document byte[] before the overall doc creation.. it will fail

[clemensutschig]

In Case One wants to Test it.. move https://github.com/opendevstack/ods-jenkins-shared-library/blob/master/src/org/ods/orchestration/FinalizeStage.groovy#L66 ahead of line 63.. then the gazillion bytes will show up