iri: fix remove_dot_segments algorithm i've misread the description in the RFC; we should copy up to, but not including, the next '/' at the end and manually inject some '/' in some other cases.

iri: fix remove_dot_segments: should copy until (including) "/"

iri: fix typo: should check the first two characters, not three

iri: fix path check path should *start* with "/" to enter that branch, not be exactly "/".

iri: fix logic in iri_unparse make sure path starts with "/" only when there's an authority part, not when it's not there!

iri: drop extraneous cpfields(IH_QUERY)

iri: set some flags as defined only when actually copying

iri: draft iri_unparse

iri: avoid unpleasant infinite loop in remove_dot_segments need to copy until the *next* path segment, otherwise we loop indefinitely if *p == '/'.

iri: catch remove_dot_segments/mergepath failures and set errno

iri: first try at remove_dot_segments and mergepath

iri: lowercase scheme and host (part of URI normalization)

draft of the new iri parser for the moment is just a URI (RFC3986) parser, and not even a complete one, but I like it much more than phos_uri. It's more gentle in the memory consumption and has a cleaner API. It's not complete, some parts are still stubbed or completely to implement, but it's the start. It compiles. Step-by-step integration of it in telescope and testing will come later.