Automatically detected by clang-tidy using performance-for-range-copy check.

This is cleaner than using exit() or assert() for that purpose.

– DataTree::packageDir() now takes a std::string_view, returns a
  std::filesystem::path, and no longer creates that directory
– DataTree::writeToFileIfModified() now takes a std::filesystem::path as
  argument
– Do not call DataTree::writeToFileIfModified() for generating MATLAB/Octave
  files, since it does not work (the directory inside which the file is written
  has been deleted by the preprocessor just before)
– Consistently use DataTree::packageDir() everywhere (for compatibility with
  planner_objective)

This is unsafe since the find() method can return a past-the-end iterator,
which should be tested for.

Replace most instances by calls to the std::map::at() method (which throws if
the key is unknown), and which is incidentally more readable.

In particular, use this feature in many loops which feature a special treatment
for the first iteration, using a boolean variable (replacing iterator
manipulation). By the way, also use std::exchange() to simultaneously test the
value of this variable and update it.

By the way, remove the unused TrendComponentModelTable::getOrigDiffVar() method.

Ref. Madysson/estimation-codes#19

This is a more natural semantics.

Incidentally, this fixes a bug in the variable mapping (M_.mapping) where some
endogenous, appearing in a log() in a VAR or TCM, would not be mentioned (e.g.
in the var-expectations/7/example1.mod test, and many others).

It contains symbol IDs, so it cannot be used from MATLAB/Octave code.

Creates a VarExpectationModelTable analogous to PacModelTable.

Ref. dynare#1837

By default, the preprocessor substitutes logs and diffs by auxiliary variables
only in equations of auxiliary models (VAR, PAC). Hence they are not
substituted in the equation defining the target, since it is outside the
auxiliary model.

The fix consists in doing the substitution in that equation as well, when we
parse the linear combination within that equation.

Ref. dynare#1837

Ref. Madysson/estimation-codes#5

Simple variable nodes are already correctly matched by
ExprNode::matchLinearCombinationOfVariables().

Also improve the related documentation.

There is no reason to keep this distinction.

Additionally, since the data structure is now symmetric with the MCE case,
unify this “h” parameter vector with the “α” parameter vector.

By the way, refactor the MCE case by merging two routines related to the Z₁ aux. var.
This restores the symmetry with the backward case, now that the latter also has
an aux. var. for the pac_expectation operator. Also store the aux. var. IDs in a
structure common to the backward and MCE cases.

Incidentally, the auxiliary endogenous variable representing Z₁ and created for
PAC MCE models no longer necessarily appear as the first auxiliary variable (so
this is effectively a revert of 64f55e4a).