Remove anytree

[valentinsulzer]

Description

It seems like we can get away with not using anytree when creating expression trees, which saves a fair amount of overhead. Functionality like render still works fine, as long as the symbol has a children attribute.
Symbol.new_copy no longer needs to be used, but I have left the functions there in case we need to go back.

Let's give it a couple of releases to be sure, but if things are working well without anytree we can also:

• remove Symbol.id and implement __eq__ and __hash__ instead (previously, these were not implemented to avoid loops in anytree)
• Simplify model.new_copy to just make shallow copies

I'll open an issue for that if this PR is approved.

Type of change

Please add a line in the relevant section of CHANGELOG.md to document the change (include PR #) - note reverse order of PR #s. If necessary, also add to the list of breaking changes.

• New feature (non-breaking change which adds functionality)
• Optimization (back-end change that speeds up the code)
• Bug fix (non-breaking change which fixes an issue)

Key checklist:

• No style issues: $ flake8
• All tests pass: $ python run-tests.py --unit
• The documentation builds: $ cd docs and then $ make clean; make html

You can run all three at once, using $ python run-tests.py --quick.

Further checks:

• Code is commented, particularly in hard-to-understand areas
• Tests added that prove fix is effective or that feature works

[valentinsulzer]

🚀

[codecov]

Codecov Report

Merging #1912 (0b02b32) into develop (b86b0c7) will increase coverage by 0.00%.
The diff coverage is 100.00%.

❗ Current head 0b02b32 differs from pull request most recent head d8c069a. Consider uploading reports for the commit d8c069a to get more accurate results

@@           Coverage Diff            @@
##           develop    #1912   +/-   ##
========================================
  Coverage    99.32%   99.33%           
========================================
  Files          346      346           
  Lines        19135    19105   -30     
========================================
- Hits         19006    18977   -29     
+ Misses         129      128    -1     

Impacted Files	Coverage Δ
pybamm/discretisations/discretisation.py	99.78% <100.00%> (-0.01%)	⬇️
pybamm/expression_tree/averages.py	100.00% <100.00%> (ø)
pybamm/expression_tree/concatenations.py	98.31% <100.00%> (ø)
pybamm/expression_tree/input_parameter.py	100.00% <100.00%> (ø)
pybamm/expression_tree/operations/jacobian.py	100.00% <100.00%> (ø)
...bamm/expression_tree/operations/replace_symbols.py	100.00% <100.00%> (ø)
pybamm/expression_tree/symbol.py	98.74% <100.00%> (-0.26%)	⬇️
pybamm/expression_tree/unary_operators.py	100.00% <100.00%> (ø)
pybamm/parameters/parameter_values.py	99.46% <100.00%> (-0.01%)	⬇️
pybamm/solvers/base_solver.py	100.00% <100.00%> (+0.49%)	⬆️
... and 3 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update b86b0c7...d8c069a. Read the comment docs.

[martinjrobins]

this looks great, glad to be rid of anytree!