feat: reintegrate workflow sub package as a part of the core

[hollandjg]

Description

Merge autora.workflow, its documentation and tests back into autora core.

• resolves feature: allow serializing/deserializing a full cycle to/from a parseable file (like YAML) autora-workflow#34
• resolves test: improve testing of serializer/deserializer with realistic StandardState data #61
• resolves refactor: merge workflow back into aurora-core #60

Type of change

• refactor: A code change that neither fixes a bug nor adds a feature
• test: Adding missing tests or correcting existing tests
• ci: Changes to our CI configuration files and scripts (i.e., those in the .github directory)

Features (Optional)

• Move the aurora-workflow code and docs back into the core.
• Merge the saving/loading tests which were defined for the state with those defined for workflow runs

Todos:

• Archive https://github.com/AutoResearch/autora-workflow
• Make a release of aurora-workflow which is dependent on autora-core < 4.1.
  https://pypi.org/project/autora-workflow/0.6.0/
• Update documentation in parent package to point to autora-core rather than autora-workflow
  • autora-core part: docs: update workflow docs #64
  • autora part: docs: update workflow documentation autora#586
• Move issue back to the autora repo - resolves Feat: Add animated cycle plotter #59
• Work out why the yaml saving and loading is slow
  • It's written in pure python. There's a libyaml version which uses C bindings, but it's not something we should rely on having. Instructions on including the CLoader and CDumper are here: https://pyyaml.org/wiki/PyYAMLDocumentation