I've always heard (and referred to it) as "Center Positive", or "Center Negative".
The image you have in the OP is "Center Positive", as the center contact is, well, positive.
To elaborate on W5VO's comment about offering to the gods. +1 by the way.
Sacrificial for protection
In my experience sacrificial component implies that the part will take some kind of damage and get destroyed in order to prevent some more precious part of the circuit from taking damage. Usually, a sacrificial part is designed so that it's easy to replace. One example, would be a common AGU fuse.
Another example. A certain instrument needs to measure an input with an expensive A/D converter. The input arrives via connector, which is exposed to the outside world. Harm can come through the connector (ESD, overvoltage, reverse polarity). A sacrificial OpAmp buffer in a socketed DIP package can be added between the connector and A/D.
http://en.wikipedia.org/wiki/Sacrificial_device
On the other hand, that all doesn't make a lot of sense in the context of O.P., in which sacrificial parts are not connected to anything. How would harm come to them? A snippet of your schematic and even a portion of the PCB layout would help understand your context better.
Sacrificial for fabrication
During fabrication* sacrificial mean that something is destroyed in the process of making the product without becoming a part of the product. Sacrificial material is a part of the fabrication process. Simple example: when you want to drill a hole, you might put a piece of wood on the other side of your part, so that the drill bit doesn't over-penetrate into something important.
* of anything, not just electronics.
May be, this is your case. May be, test points are used for some mechanical purpose. EDA package demands that they have to be connected to something (anything), so they are connected to the dummy capacitor.
Best Answer
Under that definition, in order to have electrodes a component must have nonmetallic parts. But it can't be entirely non-metallic since the electrodes are part of the component and they are metal. For example, a vacuum tube is generally made from metal, glass, and empty space (vacuum). It has both metallic and non-metallic components.
The electrodes are typically the path for current to flow in and out of the component. The exact metal used might have an effect on the performance of the part, for example in a Schottky diode.
The name of the electrode depends on the polarity: the electrode where current flows in to the component is called the anode and the electrode where current flows out is called the cathode. In the case of zener diodes, these definitions are somewhat abused.
A device with more than two electrodes (for example a transistor or a triode, tetrode, or pentode vacuum tube) necessarily has electrodes that aren't called the anode or cathode.
In vacuum tubes, there's typically an anode, a cathode, and one or more grid electrodes.
In a transistor there is no anode or cathode, just base, emitter, and collector for BJTs, or gate, drain, and source for MOSFETs.
I'm not aware of any case where that would make sense. We normally talk about magnets having poles, and electronic devices having electrodes, terminals, pins, pads, contacts, leads, etc.
In principle, this is true. In practice, we choose one terminal of a device to call the cathode and one to call the anode, based on the "normal" use conditions, and we don't change the names when the current direction changes. In the case of zener diodes, we even name the cathode and anode according to what they would be if the part were a rectifier diode, but we normally use a zener with current flowing in to the terminal we call the "cathode".