- this allows filling in the VTK structures without intermediate data
  and without sequencial insertion. Should be faster and smaller
  than the previous cell-wise insertion methods.
  Most importantly, it improves code reuse.

- this greatly simplifies data management and opens the possibility of
  reusing converted vtk meshes instead of converting each time.

- has the selected values directly and use these lookup names to store
  directly into a hash. This replaces several parallel lists of
  decomp information etc and makes it easier.

- improves the overview of the code

- avoids potentially issues if we reusing a vtkPoints array,
  and should be marginally faster without the additional
  range checking.

- adds flexiblity and reduces risk of memory leaks as we add/change
  features

STYLE: adjust naming for paraview internal polyDecomp

- easier to detect the implicit grouping

- also use updated forAll* macros

- relocated to dedicated foamVtkOutput namespace. Make it easier to
  obtain a formatter directly without a foamVtkOutput::outputOptions.
  Make the logic clear within outputOptions (avoid previous, cryptic
  bit masking). foamVtkOutput::legacy also becomes a namespace instead
  of a class. Relocate commonly used things into src/fileFormats, leave
  volField-related parts in src/conversion.

- Zero-copy does not work for several reasons, but this uses the
  OpenFOAM structures to write VTK-compatible formats into external
  arrays.

- Requires (1L << N) instead of (1 << N), otherwise it overflows
  and the result is zero.

- this should not have been there in the first place

Adds overset discretisation to selected physics:
- diffusion : overLaplacianDyMFoam
- incompressible steady : overSimpleFoam
- incompressible transient : overPimpleDyMFoam
- compressible transient: overRhoPimpleDyMFoam
- two-phase VOF: overInterDyMFoam

The overset method chosen is a parallel, fully implicit implementation
whereby the interpolation (from donor to acceptor) is inserted as an
adapted discretisation on the donor cells, such that the resulting matrix
can be solved using the standard linear solvers.

Above solvers come with a set of tutorials, showing how to create and set-up
simple simulations from scratch.

- Use on/off vs longer compressed/uncompressed.
  For consistency, replaced yes/no with on/off.

- Avoid the combination of binary/compressed,
  which is disallowed and provokes a warning anyhow

fanPressureFvPatchScalarField.H:  Correcting header documentation
RAS/TJunctionFan/0.orig: Creating tutorial for fanPressure
Adding pRef and pValue for tutorial createZeroDirectory/snappyMultiRegionHeater

Adding intertial term switch to solarLoad chtMultiRegionFoam case

- By definition, binary STL uses float (not double) when reading.
  The ascii STL should be the same. This reduces memory overhead when
  loading files. The older triSurface reader had float, the surfMesh
  reader had double, but now has float.

- Inconsistency in the STL merge-tolerances between triSurface reader,
  surfMesh reader and WM_SP vs WM_DP. Now use consistent tolerances
  conrresponding to 10,100 * doubleSMALL.

- Similar float/double code adjustments for TRI format since this is
  very similar to the STL reader and had a similar inconsistency between
  the triSurface and surfMesh version. The AC3D reader still uses
  double when reading, but this can be revisited in the future (and can
  then remove the stichTriangles method too).

- only treat text as an option if it is preceded by 0-4 spaces.
  This prevents the description of an option from being accidentally
  detected as an option.

- can avoid the intermediate point distance field in exchange for
  calculating a point subtraction twice.

Based on development contributed by Paul Edwards, Intel.

Conflicts:
	applications/utilities/parallelProcessing/reconstructParMesh/reconstructParMesh.C

    Evolves an electrical potential equation

    \f[
        \grad \left( \sigma \grad V \right)
    \f]

    where \f$ V \f$ is electrical potential and \f$\sigma\f$ is the
    electrical current

    To provide a Joule heating contribution according to:

    Differential form of Joule heating - power per unit volume:

    \f[
        \frac{d(P)}{d(V)} = J \cdot E
    \f]

    where \f$ J \f$ is the current density and \f$ E \f$ the electric
field.
    If no magnetic field is present:

    \f[
        J = \sigma E
    \f]

    The electric field given by

    \f[
        E = \grad V
    \f]

    Therefore:

    \f[
        \frac{d(P)}{d(V)} = J \cdot E
                          = (sigma E) \cdot E
                          = (sigma \grad V) \cdot \grad V
    \f]

Usage
    Isotropic (scalar) electrical conductivity
    \verbatim
    jouleHeatingSourceCoeffs
    {
        anisotropicElectricalConductivity no;

        // Optionally specify the conductivity as a function of
        // temperature
        // Note: if not supplied, this will be read from the time
        // directory
        sigma           table
        (
            (273        1e5)
            (1000       1e5)
        );
    }
    \endverbatim

    Anisotropic (vectorial) electrical conductivity
    jouleHeatingSourceCoeffs
    {
        anisotropicElectricalConductivity yes;

        coordinateSystem
        {
            type        cartesian;
            origin      (0 0 0);

            coordinateRotation
            {
                type        axesRotation;
                e1          (1 0 0);
                e3          (0 0 1);
            }
        }

        // Optionally specify sigma as a function of temperature
        //sigma           (31900 63800 127600);
        //
        //sigma           table
        //(
        //    (0      (0 0 0))
        //    (1000   (127600 127600 127600))
        //);
    }

    Where:
    \table
        Property     | Description               | Required  | Default
value
        T            | Name of temperature field | no        | T
        sigma        | Electrical conductivity as a function of
temperature |no|
        anisotropicElectricalConductivity | Anisotropic flag | yes |
    \endtable

    The electrical conductivity can be specified using either:
    - If the \c sigma entry is present the electrical conductivity is
      specified
      as a function of temperature using a Function1 type
    - If not present the sigma field will be read from file
    - If the anisotropicElectricalConductivity flag is set to 'true',
      sigma
      should be specified as a vector quantity