Frederic Tessier

Rebased on the new `add-egs-dynamic-geometry`.

Adjusted the commit titles, reordered and squashed a couple. Clean `git diff` compared to `origin/`.

Adjusted one commit message, clean `diff`.

@blakewalters @mainegra this pull request is in conflict with #974 (add dbs to radiative splitting). Since the latter is still a draft, I suggest that we merge and then resolve...

@rtownson, did "the installer get updated to include c++14 by default"?

my take on question 2: Zero the does, and this ought to be low level in `egsnrc.mortran`. If the medium is vacuum, then any dose deposition ought to be identically...

The root cause for this problem is that the B-field step size restriction is not taken into account _before_ the howfar call in the `ELECTR` subroutine when the particle is...

Thanks @victorMalkov, I was indeed wondering just now if you had dealt with vacuum transport in your implementation!

@blakewalters please double-check this pull request for merging.

- [x] Checked that `NUM_BREM` is only ever used for [information output](https://github.com/nrc-cnrc/EGSnrc/blob/e4044c1210fbbf169ee62a9452c7e176ad5367b1/HEN_HOUSE/omega/beamnrc/beamnrc.mortran#L8994-L8996), and nothing else