Paca

Replace trailing stop orders are now supported by the API.

Adding it to the SDK as well

In this release we add support for the new added feature - trailing stop:
* submit_order now supports a new type `trailing_stop`.
* 2 new optional args to submit_order: `trail_price`, `trail_percent`.

Bugs
* Fix functions related to the Watchlist endpoint.

In this release we add small changes to help the user debug issues easily:
* Indicate that websocket is connected and to which endpoint
* add a debug flag to make websocket exceptions more verbose
* explain in the docs, how to define a logger to see these changes

Other changes
* We also removed the Alpha Vantage Integration. it will be moved to a stand alone package
* We now allow working with all versions of pandas (required for Pylivetrader)

Bugs
* fix bug with polygons' historic_aggs_v2 (https://github.com/alpacahq/alpaca-trade-api-python/issues/280)
* 'NoneType' object is not iterable (https://github.com/alpacahq/alpaca-trade-api-python/issues/270)

In this release we:
* reconnect when the data websocket disconnects
* add type hints for the rest and stream modules
* add some input validators for URLs, price floats and dates
* remove deprecated polygon api wrappers: historic_trades, historic_quotes, splits
* add integration with pyup
* change the package requirements to be less strict
* update the README documentation

fixes:
* historic_agg_v2 now accepts different date formats
* adjust the sample code: martingale.py to work with the new data api

Alpaca Data API's new endpoints are supported as of this version.

- get_last_trade()
- get_last_quote()
- StreamConn() can now connect to streams of trade, quotes and minute bars provided by Alpaca streaming endpoint

Now all users (with or without live brokerage account) can access realtime quotes/trades both in REST and streaming.

As part of these changes, this release defaults streaming connection to the Alpaca streaming endpoint. If you want to keep using Polygon streaming endpoint, you will need to specify `data_stream` in the `StreamConn()` initializer, as

py
conn = StreamConn(data_stream="polygon")

For the compatibility with other entities around, we alias "start" to also be "timestamp" in the streaming bars.