This way, the execution order can be made explicit.

plugins: imap-filter-sieve: Fix mixup between tag and problem part of the FILTER response for compile error.

plugins: imap-filter-sieve: Perform main initialization of Sieve script execution environment only once.

This way, any configuration errors can be reported as a single FILTER ERRORS
response, rather than a FILTERED ERRORS response for each filtered message.

plugins: imap-filter-sieve: Make sure user gets to see at least one error when script execution fails.

The previous commit should already prevent this, but this makes sure it
can't become broken in the future either. It makes the performance a tiny
bit worse, but that's not practically noticeable.

ManageSieve doesn't allow NULs in strings.

This fixes a bug with unescaping a string with NULs: str_unescape() could
have been called for memory that points outside the allocated string,
causing heap corruption. This could cause crashes or theoretically even
result in remote code execution exploit.

Found by Nick Roessler and Rafi Rubin

Before, it only compared the handles, which would cause duplicate tests with
different values to be cached as the same duplicate test.

Also explicitly initialize login_binary->anonymous_login_acceptable to FALSE to
make clear that anonymous logins are intentionally not allowed for ManageSieve.

Addresses a Coverity report.

The action name can be determined from the sieve_action_name field that existed
before.

It is no longer used.

Avoid using a Sieve error handler directly.

lib-sieve: sieve-error - Compose both the event and the error handler log message once using the event API.

Before, it used format string and arguments.

sieve-tools: sieve-test - Use the new result message amendment callback rather than the prefix error handler.

sieve-tools: sieve-filter - Use the new result message amendment callback rather than the prefix error handler.

plugins: lda-sieve: Use the new result message amendment callback rather than a special error handler.

Before it always used the name of the action definition, which is confusing e.g.
for the keep and fileinto actions, which both map to the same "store" action
definition.

This allows logging action messages through a different (child) event.