As it is an FPGA, you can use whatever GPIOs you want for doing SPI. You are defining the SPI on a hardware level and can attach it wherever you would like (within reason).
FPGAs generally only have dedicated hardware for things like high speed serialization, microprocessor cores (see the Xilinx Zynq family or Altera Cyclone V), or memory interfaces (including dedicated memory blocks). Everything else is entirely defined by you (which is awesome).
The dedicated SPI pins are actually for configuration purposes. An SPI flash memory can be attached to those pins and the FPGA will read it when it starts up to load its configuration bitstream. The "Configuration" section in the Spartan-6 family manual talks about using SPI for initializing the device. There is usually a list of compatible SPI flash devices somewhere in the datasheets and reference manuals. The Spartan-6 is wildly popular, so I'm sure you will have no trouble finding information on those devices should you decide to use SPI for configuring your FPGA.
Short answer: You actually don't want to use those pins for your SPI interface. Those particular pins are meant for automatically configuring the device when it starts up.