combine mid and treble FFT input buffers The point of having separate input buffers for bass, mids and treble was only ever to capture lowest bass without sacrificing the responsivness of the rest of the spectrum. In the process I also made a separate buffer for treble, but as it turns out the buffer is way too small to be usefull and responsiveness is just as good with the mid buffer. So this commit removes the treble buffer and uses the mid buffer for all frquencies over the bass cut off (100Hz). The fft input buffer size is then 4096 samples for 44100 and 48000 samples rate. (double for 96k, half for 22k and so forth) although that should theoretically be able to capture frequencies down towards 10Hz it is not that good at it. It will also have a minimum 10 Hz resolution between the bars. The response time is around 100ms (or rather samples that are 100ms old will still be visualized). This will introduce some more inertia to the visualization, I have tuned down the noise filter to counter for this. The bass buffer is twice as large, and will go twice as low in the spectrum and be twice as slow. some adjustments and fixes where also made to standard eq and cut off frequencies use correct buffersizes for each band in the hardcoded eq adjust hardcoded eq to match old behaviour, less bass update cavacore blueprint, changed slightly due to adjustments