Decouple indexing from profiling

[raulcf]

EDIT: [Profiling is about 1 order of magnitude faster than indexing.
Decouple both processes]

Decouple profiling process into the smallest pieces possible. For example, one should be able to index only schemas and no data, or data alone, etc. Then find a way of combining then back again into the original form. Challenge is to maintain all perf guarantees.

[mansoure]

@raulcf how do you parallelize the indexing process? Decoupling both processes is good, but I think we may have a chance to optimize the indexing process itself. I see in another issue, #52 , that you managed to make it 3X slower instead of 10X. That is great. This might be related to issue #48 . I am also interested in helping if help is still needed.