Add a search function

A search function generates the pool of candidates that an Acquisition function then scores. Subclass BaseSearch to control where candidates come from (e.g. enumerating a dataset with DatasetSearch, or sampling from a generative model with GeneratorSearch).

Base class: BaseSearch. Key method: __call__.

Steps

1. Subclass BaseSearch under tools/alf_tools/optimizer/search/.

2. Implement __call__ to return the candidate pool for the current round.

3. Pair it with an acquisition function in an Optimizer and run a Task.

4. Add tests and run the build / pre-commit checks.

Reference notebook

• Search functions: implement a custom BaseSearch.