fix(librustzcash): bump librustzcash lib to fix CI cache issue

[borngraced]

#2292
debugging if this will actually fix CI cache issue regarding librustzcash lib

[dimxy]

Why the librustzcash revision is not tagged as 'tag' as usual, but as 'branch'

[laruh]

@borngraced is this still in progress?

[borngraced]

Why the librustzcash revision is not tagged as 'tag' as usual, but as 'branch'

I will update until after the PR is merged in librustzcash

[borngraced]

@borngraced is this still in progress?

#2299 (comment) same reason why it's in progress. I will mark ready for review when the main PR is merged in librustzcash

[onur-ozkan]

This didn't happen this often before. Something should regress this problem.. Instead of putting big patch on already heavily modified fork, can we try some hacks on this cache plugin? Did you try to play with caching key? Or at least, you mentioned that these could work too:

I think they are better than the current approach as it's just a single line addition for CI rather than making big changes on the actual code base.

[borngraced]

This didn't happen this often before. Something should regress this problem.. Instead of putting big patch on already heavily modified fork, can we try some hacks on this cache plugin? Did you try to play with caching key? Or at least, you mentioned that these could work too:

I think they are better than the current approach as it's just a single line addition for CI rather than making big changes on the actual code base.

maybe not really, and yes, while those other options could work, I actually think it makes more sense to commit the file since the original updated repo(librustzcash) now commit the file too so I guess it fine if we just delete .gitignore and commit compact_format.rs instead of "hacking"

I can try the other option for sure if you're against this.

[onur-ozkan]

I think it's still unnecessary addition. Making such changes just to bypass cache invalidation doesn't make sense IMO. Just made #2303 PR to use a rust-specific caching mechanism. Let's see if it works. If not, we can revisit this approach.