Dynamic search example not working

There seems to be a discrepancy between `intake-stac` documentation examples on [readthedocs](https://intake-stac.readthedocs.io/en/latest/installation.html) and [GitHub](https://github.com/intake/intake-stac/blob/main/docs/source/tutorial.rst). The example explaining how STAC items are searched based on a query is different between those sources, where the former works with `satsearch`and the latter with `pystac-client`. Perhaps the former is outdated or the latter merely a development version. However none of these examples seem to work with the following package latest versions:

``` python
import satsearch
print(satsearch.__version__)
```

0.3.0

``` python
import intake
print(intake.__version__)
```

2.0.3

``` python
import intake_stac
print(intake_stac.__version__)
```

0.4.0

``` python
import pystac_client
print(pystac_client.__version__)
```

0.7.6

## `satsearch` example

``` python
URL='https://earth-search.aws.element84.com/v0'

results = satsearch.Search.search(
url=URL,
collections=['landsat-8-l1-c1'],
bbox=[43.16, -11.32, 43.54, -11.96]
)
items = results.items()
catalog = intake.open_stac_item_collection(items)
list(catalog)
```

ValueError: Expected instance, got:

## `pystac_client` example

``` python
import pystac_client
URL = "https://earth-search.aws.element84.com/v0"
catalog = pystac_client.Client.open(URL)

results = catalog.search(
collections=["sentinel-s2-l2a-cogs"],
bbox = [35.48, -3.24, 35.58, -3.14],
datetime="2020-07-01/2020-08-15")

items = results.get_all_items()
catalog = intake.open_stac_item_collection(items)
list(catalog)
```

DoesNotConformTo: Server does not conform to ITEM_SEARCH, There is no fallback option available for search.

Compare the examples in docs/source/tutorial.rst and the Read the Docs installation page, then reproduce both snippets with the package versions listed in the issue. Trace the failing search and item-collection paths before updating the documentation so the examples use one supported, working approach and agree across both pages.

Written by the indexing model from the issue text.