Re: [PATCH] builtin-merge: fail properly when we are in the middle of a conflicted merge

We've exchanged quite a few "here is a better one", "oops, that is not
enough", which is confusing to bystanders.  Here is my proposed final
series, meant to be applied to 'maint'.

  [1/2] merge: fix numerus bugs around "trivial merge" area
  [2/2] unpack_trees(): protect the handcrafted in-core index from read_cache()