Prints summary of properties for each task, selecting only a single row for each unique status/property/task. Prefers the earliest FAIL or the latest non-FAIL, using the same formatting as the `--livecsv`.
`SbyTask::update_status()` optionally takes the current step/depth, which gets used for some solvers/engines when exiting to pass unknown properties.
btor engine tracks the current step, but it doesn't track/report which property triggered a CEX so it's only useful on exit.
Use data source as a fallback if engine isn't provided (such as when coming from the `task_status` instead of an engine update).
Currently unconditional, and only for aiger and smtbmc.
Uses task.name from intertask cancellation.
Stores `time.time()` when calling `SbyStatusDb::create_task()` in order to calculate time since start of task (consider reducing to integer seconds).
Always call `_setup()`, but use `CREATE TABLE IF NOT EXISTS`. The sql schema doesn't include this, so inject it during the setup instead of adding it to the `SQLSCRIPT`.
Requires python >= 3.11 (oss-cad-suite is 3.11.6, but there isn't any minimum python version given for SBY that I can find).
Makes the error less opaque (even though it still has the trace), which I think is necessary given that using the status db is totally optional and otherwise using a different version of SBY with an extra db field in a directory where a previous version has run will just crash with an obscure sqlite3.OperationalError.
`--status` reports if the schema doesn't match.
`--statusreset` is able to perform a hard reset, dropping all the existing tables and calling `_setup()`.
Use context manager to handle commit/rollback.
Use `sqlite3.Connection.in_transaction` property instead of rolling our own.
Read-only methods don't need the transaction wrapper and we never read-update-write.
Status directories are normally ignored because they have a sqlite file, but it's possible to create a status dir without a database when using `--setup`.
