Non allocating R2N and R2DH

[MaxenceGollier]

No description provided.

[MaxenceGollier]

@dpo @MohamedLaghdafHABIBOULLAH. Should be merged after #159 and #164.
Also, there is a small issue when comparing R2N with R2DH as a subsolver with the older version of #153, the results are very slightly different, I am still working on it.
I tested locally that there were no differences in the other case. I will show this here tomorrow.
Finally, the allocation tests will pass as soon as opnorm does not allocate and #164 is merged.

Other than that, let me know what you think.

Fyi @nathanemac.

[MaxenceGollier]

After discussing with @MohamedLaghdafHABIBOULLAH, there is an error in the original version which explains the differences with R2N called with R2DH.
I think we can start to review this PR @dpo though I need #164 to be merged first for this to work (and in order to show the benchmarks and test that the results are the same with the older version)

[MaxenceGollier]

This has too many conflicts, I am opening a new PR...