Feature suggestion

It would be great to be able to create an extraction pipeline based on examples and prompt once...
`my_custom_extractor = lx.extract(prompt=prompt, examples=examples)`

And then reuse that pipeline across multiple input texts...
```text1 = "The person's name is Anne Teak, and they are 45."
result1 = my_custom_extractor.extract(text1)
print(result1)
```

and then...
```text2 = "The person's name is Scott Chegg, and they are 61."
result2 = my_custom_extractor.extract(text2)
print(result2)
```

I would imagine this would reduce the work required / speed up each separate extraction?