Output goes full scale if not continuously fed with samples

[ed-xmos]

When PWM halts, we get one or zero at the output.
It is likely hard to fix this in the filter stage due to low MIPS @192KHz but the SD/PWM stage might be able to accomodate this and output a 50% duty PWM.

[henkmuller]

I wonder whether to tackle this at the same time as issue #21

They are two sides of the same coin; the problem is that when for some reason the sample generating thread pauses it both drives a DC level and the two channels go out of sync.

One way to solve them is to run this thread asynchronously, ie, it outputs samples out of a circular buffer, whatever happens; and it stores where it is so that when the other thread has a hiccup it realigns itself 180 degrees out of phase. That does rely on the other thread knowing that a hiccup has been resolved.

Looking forward to other ideas.

[ed-xmos]

Addressed by #17 which adds a select which times out to an alternating idle word.

This is a last resort as any normal stream start/stop should be carefully muted by the app.