adding complete_adapt_stream algorithm

[lehecka]

similar to next_adapt_steram, adding complete_adapt_stream

[MarkCMann]

Could this instead be written as using the existing adapt_stream algorithm but passing the identity function for the adapt_next param? That would make this algorithm more maintainable in the long run.

[lehecka]

I used the next_adapt_stream implementation as a template for this algorithm. I can see how both next_adapt_stream and complete_adapt_stream can be written on top of adapt_stream. I am assuming this would be more optimal implementation since it saves some bytes for storing the transformation lambda for the other signal which is not used.

[MarkCMann]

If the function is just an identity transformation then it shouldn't take up any additional space in the adapt_stream type (so long as the functions are annotated with [[no_unique_address]]). Check out this godbolt for an example of how wrapping an identity transform function doesn't require additional storage in the wrapper class.