OK, with the clarifications, it seems that the DCC decoder in the locomotive is sending the motor the same polarity no matter what you intend. That decoder is between the electric motor in the locomotive and the track, with a bridge rectifier between the track and the decoder. So, no matter whether there is AC or DC on the track, and, if DC, no matter which rail is + and which is -, the decoder always sees the same DC power polarity as its input. It needs to determine which wire from the decoder to the electric motor gets the + and which gets the minus, based on programming in the decoder and maybe programming in the MRC controller. It is possible that the 2 are not completely compatible, or maybe one or the other is malfunctioning.
I really don't understand how the MRC Tech 6 is designed. It would help to know what is really happening when you push the direction buttons on the controller. Do you have a voltmeter? (You really do need one to work in this hobby, but it can be a simple. cheap one - you don't need a special DCC meter, even when using DCC.)
If/when you have a voltmeter, measure the voltage across the rails as you increase the throttle. In a normal DC controller, it should go from zero volts to about 12 volts on the DC scale. With it somewhere in the 10 to 12 volt range, push the direction buttons and see if the polarity on the rails changes. For a regular DC controller, the rail that is + should become - and the rail that is - should become + to change the direction of a DC electric motor that is directly connected to the rails. Knowing whether your controller is actually doing that will help us to figure out what is happening. Most DC controllers use a simple dp/dt slide switch to change the rail polarity. Because your controller has buttons and is designed to send some DCC sound signals to sound decoders, I am suspecting that those buttons might do things differently.
Another question is whether you bought the locomotive second hand. If so, it is possible that the previous owner had changed a CV (configuration variable) in the decoder to not run on DC. A lot of people who run DC/DCC capable decoders turn off the DC capability when they run them on DCC systems to prevent full speed runaways that sometimes happen when a decoder doesn't see a DCC signal soon enough after it gets power, so it assumes it is on a DC system, and, because it is really on a DCC system, it sees full track power and sends that to the electric motor. Without the ability to use a DCC controller to read the appropriate CV in your particular decoder we are going to have a hard time figuring that out.
Have you joined the MRC discussion group on Groups.io? That is the place that is most likely to be able to give you advice on what your controller is actually doing, and whether they have seen this problem before and know how to fix it.
I really don't understand how the MRC Tech 6 is designed. It would help to know what is really happening when you push the direction buttons on the controller. Do you have a voltmeter? (You really do need one to work in this hobby, but it can be a simple. cheap one - you don't need a special DCC meter, even when using DCC.)
If/when you have a voltmeter, measure the voltage across the rails as you increase the throttle. In a normal DC controller, it should go from zero volts to about 12 volts on the DC scale. With it somewhere in the 10 to 12 volt range, push the direction buttons and see if the polarity on the rails changes. For a regular DC controller, the rail that is + should become - and the rail that is - should become + to change the direction of a DC electric motor that is directly connected to the rails. Knowing whether your controller is actually doing that will help us to figure out what is happening. Most DC controllers use a simple dp/dt slide switch to change the rail polarity. Because your controller has buttons and is designed to send some DCC sound signals to sound decoders, I am suspecting that those buttons might do things differently.
Another question is whether you bought the locomotive second hand. If so, it is possible that the previous owner had changed a CV (configuration variable) in the decoder to not run on DC. A lot of people who run DC/DCC capable decoders turn off the DC capability when they run them on DCC systems to prevent full speed runaways that sometimes happen when a decoder doesn't see a DCC signal soon enough after it gets power, so it assumes it is on a DC system, and, because it is really on a DCC system, it sees full track power and sends that to the electric motor. Without the ability to use a DCC controller to read the appropriate CV in your particular decoder we are going to have a hard time figuring that out.
Have you joined the MRC discussion group on Groups.io? That is the place that is most likely to be able to give you advice on what your controller is actually doing, and whether they have seen this problem before and know how to fix it.