Feature 7/2 Katana Runner Macro

[matzayonc]

In this PR the focus laid on overcoming limitations of benchmarks.

• need for a katana instance running in the background - by katana runner
• need for a manual deploy to the running instance - by deployer and sozo
• excessively long test run times - by static runner in each proptest test
• log reading dificulty - by creating a separate log file for each test, with name derived from the tests name
• test boilerplate - by employing a procedural macro to generate all needed
• non-async nature of proptest tests - by initializing a temporary runtime in a macro
• nonce fetching delays - by generating fresh accounts for each test or sending only estimate transactions