The strip_comments function uses the count of quotes to know if a
comment char (';' or '#') is the start of a comment or part of the
multiline as a string.
Unfortunately converting the count of quotes from previous lines to a
boolean meant that it would only work as expected in some cases
(0 quotes or an odd number of quotes).

Instead of a `git_bom_t` that a `git_buf` function returns, let's keep
it `git_buf_bom_t`.

The `git_buf_text` namespace is unnecessary and strange.  Remove it,
just keep the functions prefixed with `git_buf`.

Signed-off-by: Sven Strickroth <email@cs-ware.de>

Right now, all configuration file backends are expected to
directly mess with the configuration parser's internals in order
to set it up. Let's avoid doing that by implementing both a
`git_config_parser_init` and `git_config_parser_dispose` function
to clearly define the interface between configuration backends
and the parser.

Ideally, we would make the `git_config_parser` structure
definition private to its implementation. But as that would
require an additional memory allocation that was not required
before we just live with it being visible to others.

The config parser code needs to keep track of the current parsed
file's name so that we are able to provide proper error messages
to the user. Right now, we do that by storing a `git_config_file`
in the parser structure, but as that is a specific backend and
the parser aims to be generic, it is a layering violation.

Switch over to use a simple string to fix that.

By convention, parameters that get passed to callbacks are
usually named `payload` in our codebase. Rename the `data`
parameters in the configuration parser callbacks to `payload` to
avoid confusion.

The `parse_section_header_ext` name suggests that it as an extended
function for parsing the section header.  It is not.  Rename it to
`parse_subsection_header` to better reflect its true mission.