cargo: Update sp-* crates from v7.0.0 to v9.0.0

[lexnv]

Update the sp-core, sp-keyring and sp-runtime crates to the latest v8.0.0 version.

The crates need a manual update to properly sync all dependencies, as noted by
dependabot in #747, #746 and #745.

[niklasad1]

@lexnv looks like the build script in wasmtime-runtime fails, is there something weird going with our C toolchain in the CI?

[lexnv]

Yep, there seems to be an issue with the wasmtime-runtime CI step. It seems that there is one file missing needed for compilation, it might be possible that is not included in the environment path:

warning: src/helpers.c:1:10: fatal error: 'setjmp.h' file not found
warning: #include <setjmp.h>

[niklasad1]

I just checked that wasmtime-runtime wasn't part of the dependency tree before and it seems not compile for WASM.
(I could reproduce it locally as well)

https://crates.io/crates/sp-wasm-interface/8.0.0/dependencies is the cause that wasmtime-runtime is brought into the dependency tree but I don't follow we are using no-default-features = true is it a bug in substrate?

[niklasad1]

wasmtime is hard-dependency in substrate now which causes the wasm stuff to not compile anymore...

see https://github.com/paritytech/substrate/pull/12684/files#diff-c2654579a98c27b848ec78860417cbec4375ed7c1dc8df990d77c6ea6f68428cR26

[jsdw]

Ok, so summary so far:

• Removing wasmtime means disabling the "std" feature (ie default-features = false).
• But disabling this features in Substrate makes it assume we want a WASM build so a bunch of other stuff goes wrong.
• I think it's worth experimenting with removing our dependencies on substrate crates and see how much of a pain (or not) it'll be to take what we want into Subxt itself.
• I'm sortof keen on this anyway if we can do it without copying too much code (ie just rely on hashing libs directly and make some traits on our side that it's fairly easy to impl for substrate things etc). It may let us streamline and simplify stuff.

[jsdw]

I'll close this now since it's been superceded by #760