Ensemble Object

Nobody has claimed this yet.

When working on an application, I wanted a method to easily combine multiple sampling strategies, one idea was to have an `Ensemble` object which matches the API of the sampler objects, but has multiple samplers. Partial pseudo-code below:

```
class Ensemble:
"""
Class for multiple samplers.
"""
def __init__(self, sampler_data):
if not isinstance(sampler_data, list):
sampler_data = [sampler_data]
self._samplers = [new_sampler(data) for data in sampler_data]
def check_validity(self):
for sampler in self._samplers:
sampler.check_validity()
# check that all samplers have the same parameter names, etc.
def get_samples(self):
samples = []
for sampler in self._samplers:
samples.extend(sampler.get_samples())
return samples
```

Then you could specify multiple samplers via the yaml interface:

```
sampler:
- type: list
parameters:
X1: [5, 10]
X2: [5, 10]
- type: cross_product
parameters:
X1: [15, 20, 30]
X2: [15, 20, 30]
```

No implementation files or tests are named. Start by locating the existing sampler classes and the YAML sampler-parsing entry point, then compare their validation and sample APIs. Done means multiple configured samplers can be combined through an Ensemble-like API and their samples are returned together.

Written by the indexing model from the issue text.