fix: Do not use nHeight when trying to identify the very first/initial snapshot

[UdjinM6]

Issue being fixed or feature implemented

-1 will only mean not initialized from now

Should fix crashes like https://gitlab.com/dashpay/dash/-/jobs/4893359925
This fix applied on top of #5525 https://gitlab.com/UdjinM6/dash/-/pipelines/972154075

What was done?

Introduce and use m_initial_snapshot_index instead of re-using nHeight. Added a couple of asserts to make sure:

1. we never create mn lists with nHeight set to -1 explicitly (but it's ok for ctor with no params to do so)
2. we never set nHeight to -1 for an existing mn list
3. we never try to get a height for a non-initialized list
4. GetListForBlockInternal never returns non-initialized mn lists

How Has This Been Tested?

Run tests, run regtest/testnet wallets.

Breaking Changes

We never stored snapshots with nHeight == -1, should be no breaking changes I think.

Checklist:

• I have performed a self-review of my own code
• I have commented my code, particularly in hard-to-understand areas
• I have added or updated relevant unit/integration/functional/e2e tests
• I have made corresponding changes to the documentation
• I have assigned this pull request to a milestone (for repository code-owners and collaborators only)

[knst]

concept ACK.
It seems as QT's callback from Timer indeed can be called before blockchain's data fully initialized.

[UdjinM6]

That's what I thought initially too but looking at logs a bit more - node1 was at height=14 already when it crashed, it must be some deeper issue.

[ogabrielides]

ACK

[UdjinM6]

I think I found the root of it (pls check PR description).

[ogabrielides]

LGTM

[PastaPastaPasta]

utACK for squash merge

[knst]

LGTM