I am developing an iOS game that will be played with real money. The problem that I am facing is the security behind the game. First of all, the game is client-side. When the player buys into a game, a call is made to my REST API. This call logs a game transaction in MySQL and deducts from the players balance. When the game is finished, the client makes a second call to the API. This second call updates the previous transaction with new information such as a completion timestamp, cashout amount, and status. If the player won, their balance will be credited.
I know that as a developer you cannot trust the client. However, I'm at a standstill with this. I have almost the entire infrastructure coded from scratch. This infrastructure includes a cashier system, user management system, accounting system, Python multiplayer game server (also controlling cash flow), and an iOS Objective-C framework to interact with the web service. All made from scratch over a long period of time.
Yet, the client-side security is holding me back. I'm asking for any advice on what to do in my position. I can't just have someone get their API key, find out how to call the web service, and setup a fake transaction and win.