* Merge configurable fields with previous checkpoint config.
* Update update_state() and aupdate_state() to merge configurable fields with previous checkpoint config.
* Update tests to verify that all checkpoint metadata contain the expected configurable field keys. This assertion is needed because a run can have an arbitrary number of steps based on the construction of the graph.
- this avoids conflicts if multiple processes creating checkpoints in same thread at same time
- uuid6 with a monotically increasing clock_seq is sortable by creation time at ms precision (plus clock_seq for ties, plus 48 bits of randomness for further ties)
- Remove pydantic usage from base checkpointer class
- Make serialization configurable for all existing checkpointer classes, you can eg use dill or json instead of pickle
- Add Pregel.get_state_history and .aget_state_history methods to get history iterator
- Update checkpointer base class with new list and get_tuple methods
- Rewrite in-memory checkpointer class to track history
- Rewrite sqlite and aiosqlite checkpointers to track history
- Add new tests for history tracking
- checkpoints are now copied before being mutated
- compiled graph and compiled state graph no longer need to override get_state/update_state
- pregel class now natively supports interrupt before/after
