This is an interesting question, but I think the basic answer is "no". The lowest level at which symbols are applied is for a geoelement (feature or graphic, ie), so it sounds like your approach of showing a larger geometry using geoelements for each segment is the best workaround. If you only want to display feature segments individually but not store them that way, you might consider a "hybrid" approach where you store only the larger feature in your dataset, but use a graphics overlay to display the segments with unique symbols. Especially if you only need to display segments for certain workflows. Perhaps allow the user to select a polyline that they want to see the segments for (?).
Performance might be a concern if you start handling lots of segments (thousands). The guide doc topic here has some tips for improving performance when working with a graphics-intensive app: https://developers.arcgis.com/net/programming-patterns/performance-considerations/#graphics
One tip is to use a renderer for a graphics overlay rather than symbols for each graphic. To implement that, you might do something like generate a random number for each graphic (perhaps between 1-10, eg) and apply a renderer to the overlay that shows a unique color for each possible value (1-10, or whatever range you use).
