SCM's ccpp-physics repo is moving to scm/dev branch

[scrasmussen]

SOURCE: Soren Rasmussen, NSF NCAR

DESCRIPTION OF CHANGES:

• SCM will use a scm/dev branch in the NCAR/ccpp-physics repo
  • It has been agreed that the NCAR/ccpp-physics is moving to a new code management style to increase the pace of development. This will require a scm/dev branch to be created from NCAR/ccpp-physics's main. This branch will be as up-to-date as possible with ccpp-physic's main and allow SCM to have a functioning build while it digests new PR's from ccpp-physics
• This PR points to @scrasmussen ccpp-physics scm/dev branch
  • The branch was created with git revert 85f313462b0d410df2ff823e7f030e11d949a833 to revert agreed-upon breaking changes in Read and broadcast data from MPI root rank during init and timestep init phase in GFS time vary; add time vary interstitials for NEPTUNE ccpp-physics#1187
  • This is a know issue that @grantfirl is working on, once SCM can build the Read and broadcast data from MPI root rank during init and timestep init phase in GFS time vary; add time vary interstitials for NEPTUNE ccpp-physics#1187 changes, that PR will be added back into the scm/dev branch by reverting the revert commit with git revert c8d1e9c604af9309abd8521332e764cca2a70a4e
• scm/dev branch needs to be created from NCAR/ccpp-physics's main. Wanted to make sure everyone agrees with this process before doing it. Once that branch is created a PR with scm/dev branch will be created

ASSOCIATED PRs:

• @scrasmussen's scm/dev branch will be PR'ed into scm/dev once that is created and this bullet will be update to list that PR number

TESTS CONDUCTED: Built all suites with GNU on Derecho and ran twpice case

[grantfirl]

@scrasmussen I just created the branch of scm/dev in the NCAR fork from the latest commit that the SCM is using. This leaves out the mpi bcast PR merge and the IP merge, for which there is a SCM PR for already. I don't think that you need your scm/dev branch in your fork.