SetVoiceStack looks for "!v" in variant_name and skips the first
three characters if "!v" is found. The problem here is that it
does not check that the third character is the path separator, so
may advance into unknown memory if variant_name is exactly "!v".

This fixes that problem by checking for the path separator. It
also simplifies the logic by checking the bytes explicitly.

NOTE: This is not strictly needed, as the only code paths this is
relevant for is in espeak_ng_SetVoiceByName, and the variant name
comes from ExtractVoiceVariantName, which sets up the variant name
correctly.

Various places call SetVoiceStack with "" for the variant_name. This
causes -fsanitize=address to fail with an overflow as the call to
memcmp is checking the first 2 bytes, and there is only 1 byte
available.

compiledict.c sets dict_name to dictionary_name if dict_name is
not set, and passes that to LoadDictionary. LoadDictionary then
copies the passed in name to dictionary_name.

This causes -fsanitize=address to fail with overlapping memory
addresses passed to strncpy (copying the string to itself). As
such, don't copy the name in this case.

Z-SAMPA is a more complex specification than Kirshenbaum, X-SAMPA,
and CXS. Additionally, it specifies phonemes such as palatal and
velar trills that are not defined by IPA and in the context of the
velar trill is marked as being impossible.

As such, Z-SAMPA will not be supported in eSpeak NG for now. That
may change in the future once support for the other transcription
schemes has been implemented.

Fixes two sanitizer warnings:

```
src/libespeak-ng/compiledata.c:2291:27: runtime error: left shift of 1 by 31 places cannot be represented in type 'int'
src/libespeak-ng/compiledata.c:2424:34: runtime error: left shift of 1 by 31 places cannot be represented in type 'int'
```

This makes it clearer that both variant and private-use tags are
used.

According to page 5 of [RFC5645](https://tools.ietf.org/html/bcp47#page-1-5),
variant subtags must have 5 to 8 characters. A 9-character subtag is
sytactically ill-formed and should be rejected by conforming parsers.

Technically, BCP47 does not allow the use of unregistered variant subtags.
For `fonkirsh`, I’ve filed a
[language variant subtag registration request](https://mailarchive.ietf.org/arch/msg/ietf-languages/1IrLl3n4wJ4Fr1xXV34QquJn8Bc)
with the IETF. If IETF rejects my request, we could probably add
Kirshenbaum to Unicode’s registry for [RFC6497](https://tools.ietf.org/html/rfc6497).

Not sure what to do about the other non-registered variants. If they’re
commonly used, it might be an option to register them as well, but
they seem to be even more exotic than Kirshenbaum, and we wouldn’t to
bloat the BCP47 registry too much. If espeak-ng really wants to handle these
alphabets, it would probably make sense to use BCP47 private-use extensions,
such as `x-fonzsamp`. According to IETF BCP47, private-use subtags can have
up to (but not more than) 8 characters, so I’m shortening these too.

This affect e.g. pronunciation of я (ja) because j then starts from
pause

This reverts commit 39994150ba.

No modification was being done to buf after the copy of filepath.
It was just being passed to fopen. As such the copy is redundant,
and can lead to buffer overflow issues if the specified filepath
is larger than buf.