Remove append qmc section input attribute and attempts to implement from source.

**Is your feature request related to a problem? Please describe.**
Remove parsing of this input option and the untested and attempts at implementing it. The semantics of this attribute and its interaction with qmc section inputs is difficult to reason about and its proper functioning simulation time has inadequate coverage.

I think the motivation here was to just have the stat.h5 scalar.dat just continue on with the next section. I guess that would make plotting easier but I think this is a data analysis convenience complicating the application code.

To keep it we need to determine the following:
How do you determine a qmc section is valid input when this turned on? It's much more complex and stateful now. How do we produce "valid" self contained hdf5 files, now you have to somehow indicate that meta data may only apply to part of a hyperslab of data.

**Describe the solution you'd like**
Remove all traces of the qmc section append feature.

**Describe alternatives you've considered**
Don't do anything. The documentation in the manual has at least three ?'s which should scare of the sensible.

**Additional context**
Add any other context or screenshots about the feature request here.

Locate the qmc section append input parsing, its implementation, and the related manual entries; the issue does not name specific files or tests. Remove the feature's source and documentation traces, then verify that no append-related input handling remains and that existing tests still pass.

Written by the indexing model from the issue text.