remove snapshot_after_ops configurable, auto-tune threshold for size and time

[spacejam]

we should shapshot when:

• unmerged heap pointers use up a lot of space relative to the size of the last written snapshot
• the log would take a long time to read during recovery due to the last snapshot being outdated

we should not snapshot when:

• the snapshot thread has been active above a certain % of the time (ignored or reduced during testing)