In this challenge, you will apply your knowledge of linear regression and data visualization to forecast future inflation rates. You will work with a hardcoded pandas DataFrame containing annual inflation rates for a country over the past 15 years. Your goal is to write a function that fits a linear regression model to this data and then uses it to predict the inflation rate for the next two years. Additionally, you will visualize both the historical and forecasted inflation rates using matplotlib to clearly communicate the results.

To begin, consider how economic forecasting often relies on historical trends to estimate future values. By fitting a linear regression model to past inflation data, you can capture the underlying trend and use it to make informed predictions. Plotting the results allows you to compare the actual historical data with your model's forecasts, which is a common practice in economic reporting and analysis.

You will need to use pandas to manage the data, scikit-learn's LinearRegression to fit the model, and matplotlib to create the visualization. The function you write should take no arguments, operate on the provided DataFrame, and return the two forecasted inflation rates as a tuple. The plot should show the historical inflation rates as well as the regression line extended to include the two forecasted years.