Isovec

- added `data` property to `Substance` to store additional information or to calculate mean values of mixtures
- updated `copy()` function and documentation of all constructors
- added `rule_of_mixtures()` method to `Mixture` class
- used to calculate mean values of keys in `data` dictionary
- all constituents of a mixture must have this key-value pair
- fraction mode for weighting must be chosen: either atomic/mole, weight or volume fractions
- can use general or inverse rule of mixtures, cf. https://en.wikipedia.org/wiki/Rule_of_mixtures
- changed calculation of density in `_calc_rho()` of class `Mixture`:
- calls new `rule_of_mixtures()` method with special "density" keyword
- now uses volume fractions and general rule of mixtures (instead of old direct implementation with weight fractions and inverse rule)

- added `copy()` function to `Substance`
- deep-copy of the substance
- keyword arguments to change properties are:
- `name` for the descriptive name
- `symbol` for the symbol
- `rho` for density
- `M` for molar mass

- fixed typos in "README.md"

This update features a major overhaul of the routines to fetch isotopic (and elemental) compositions.
Atomic fractions of `Molecule`s in `Mixture`s cannot be passed down directly.
This would underestimate the atomic fraction of the `Element`s of that `Molecule`.
Instead, all `Element`s have to be collected according to their actual atomic abundance and subsequently be normalised.
When the composition input mode was `weight` and the isotopic composition was fetched in `weight` as well, values were **correct** (because of back and forth conversion).
Update 1.1.1 was ment to adress the faulty isotopic composition, when fetched in ´atomic´ mode, but introduced other errors and is thus discouraged to be used.
The respective releases on PyPI therefore have been yanked.
This new approach also required changes to the tree-printing routine.
`print_tree()` has been replaced by `print_tree_input()`, that acts like the old `print_tree(scale=False)`.
To show the actual (scaled) elemental and isotopic composition, `print_tree_composition()` has to be used.

- added `get_elements()` method to `Substance` for elemental composition of that `Substance`
- keyword `mode` for either "atomic" or "weight" fractions
- added option for `get_isotopes()` to supply explicit collection of (natural) elements to be considered
- removed the internal method `_append_isotopes()`
- instead uses `_isotopic_composition()`, `_elemental_composition()` and `_append_elements()` to get correct (atomic) composition
- `_isotopic_composition()`, `_elemental_composition()` return a dictionary, matching the position in the tree hierarchy to the `Isotope` or `Element` and their fraction respectively
- added `surrogate_isotope()` method to `Element` to get a surrogate `Isotope` object in case of a natural element (used for `use_natural` algorithm)
- removed `print_overview()` from `Mixture`, `Molecule` and `Element`, because it doesn't comply with the new approach and has been superseded by tree-printing
- `Node`s in the tree use their position in the tree-hierarchy as their id
- replaced `print_tree()` with
- `print_tree_input()`
- shows input composition in selected fraction modes (`atomic`, `weight` and `volume`)
- similar to old `print_tree(scale=False)`
- `print_tree_composition()`
- shows scaled composition of `Element`s and `Isotopes`s of a `Substance`
- possible fraction modes: `atomic` and `weight`
- uses `Node` id to fetch the correct composition value from `_isotopic_composition()` or `_elemental_composition()` dictionary
- added more sophisticated right alignment options for tree-plotting
- added `_print_depth()` method for `Node` class, to get desired level of printing that `Node` object
- new prefix generation in `print_node()` of `Node` class
- right-alignment-preference class variable for `Substance`
- `Element`s and `Isotope`s are respectively right aligned in `print_tree_composition()`
- added internal conversion consistency case to "validation.py"
- added detailed air case to "validation.py"

- fixed problem with tree plotting of molecules, that lead to wrong atomic isotope vectors
- changed molar mass of molecule to average molar mass (as other substances)
- added molar mass of molecule (is calculated as before)
- added `_get_data_dict()` to `Substance` (and `Molecule`) for more flexibility in tree plotting
- added option for `get_isotopes()` to supply explicit collection of (natural) elements to be considered

- fixed problem with molar mass of molecules, that lead to wrong atomic isotope vectors
- changed molar mass of molecule to average molar mass (as other substances)
- added molar mass of molecule (is calculated as before)
- added `_get_data_dict()` to `Substance` (and `Molecule`) for more flexibility in tree plotting
- added option for `get_isotopes()` to supply explicit collection of (natural) elements to be considered