Update old RAKP formats for consistency with new ones

[solardiz]

After merging #6005, we may want to rename our old RAKP formats (CPU and OpenCL) to include SHA1 in the name for consistency with the newly added formats. We may also want to drop FMT_HUGE_INPUT from them.

[solardiz]

I've just tried renaming the formats and dropping FMT_HUGE_INPUT, but for the CPU format I'm getting:

Benchmarking: RAKP-SHA1, IPMI 2.0 RAKP (RMCP+) [HMAC-SHA1 512/512 AVX-512 16x]... (8xOMP) FAILED (FMT_HUGE_INPUT should be set)

Somehow the OpenCL format seems fine without FMT_HUGE_INPUT.

Oh, and besides these two kinds of changes, we could also want to add the binary hash functions like the new formats have.

[solardiz]

Somehow the OpenCL format seems fine without FMT_HUGE_INPUT.

So that's because of salt size difference - CPU format uses a struct and its size, OpenCL one directly computes a size from constants. If we define the binary hash functions, then this requirement will be skipped.

        if ((!format->methods.binary_hash[0] || format->methods.binary_hash[0] ==
             fmt_default_binary_hash) && format->params.salt_size > 512 &&
            !(format->params.flags & FMT_HUGE_INPUT))
                return "FMT_HUGE_INPUT should be set";

I think I'll commit the rename, but preserve FMT_HUGE_INPUT for now - maybe @magnumripper will make the proper changes later.