Adjust internal socket API for afpacket / afring to support no-limit BPF filter option

[fako1024]

Restructures the internal API for the socket options into a struct{} based one (too many options in the method signature anyway) and hence allows exposure of a new option function for afring / afpacket sources:

listener, err := afring.NewSource(devName,
	afring.CaptureLength(filter.CaptureLengthFixed(64)),
	afring.BufferSize((1<<20), 4),
	afring.Promiscuous(false),
	afring.IgnoreVLANs(ignoreVLANs),
	afring.DisableAutoBPF(),                   // Additional option (no breaking API change on this level)
)

Using this there will be no BPF filtering introduced beyond the flag-based ones (i.e. snaplen, promiscious mode and VLAN ignore mode).

Extra BPF instructions can still be provided in this mode, effectively providing the option to set arbitrary BPF from scratch.

Closes #96