[FEA]Vamana API Disk Index Support

[jperez999]

Is your feature request related to a problem? Please describe.
The Vamana python api currently only supports getting loaded by StaticMemoryIndex of diskannpy,
need the Vamana api exposed via python to support exporting the index in a way that can be used by the StaticDiskIndex class in diskannpy (https://microsoft.github.io/DiskANN/docs/python/latest/diskannpy.html#StaticDiskIndex).
We need this to be able to allow customers to use the disk index so that they can support very large indexes that might not fit on resource (cpu/gpu). So we want to allow them to use it from disk, even if search might be slower.

Describe the solution you'd like
I'd like to see the saving of the vamana graph in cuvs work with the vectors_from_file (https://microsoft.github.io/DiskANN/docs/python/latest/diskannpy.html#vectors_from_file), so that I can then create the appropriate StaticDiskIndex.

[tarang-jain]

I think StaticDiskIndex should work too with the correct parameters for serialization. I'll try to post an example next week.

[jperez999]

OK cool, from what I saw in the function documentation in the code, it seems that the index needs to have certain files:

        ### Parameters
        - **index_directory**: The directory containing the index files. This directory must contain the following
            files:
            - `{index_prefix}_sample_data.bin`
            - `{index_prefix}_mem.index.data`
            - `{index_prefix}_pq_compressed.bin`
            - `{index_prefix}_pq_pivots.bin`
            - `{index_prefix}_sample_ids.bin`
            - `{index_prefix}_disk.index`

So I guess I am using the wrong options when I go to save the index in cuvs. Look forward to the example!