Add documentation on building, testing, benchmarking (#352)

Author: sjakobi

CONTRIBUTING.md

@@ -0,0 +1,60 @@
+# Contributing
+
+## Building, testing, benchmarking
+
+Building the library:
+
+```
+cabal build
+```
+
+Running the testsuite:
+
+```
+cabal test
+```
+
+Viewing the test options:
+
+```
+cabal run tests -- --help
+```
+
+Running a specific property test with an increased number of test cases
+(default: 100 cases):
+
+```
+cabal run tests -- -p '/All.Properties.Data.HashSet.basic interface.member/' --quickcheck-tests 100_000
+```
+
+Running the benchmarks:
+
+```
+cabal bench
+```
+
+Viewing the benchmark options:
+
+```
+cabal run benches -- --help
+```
+
+Running a specific benchmark with a reduced target standard deviation (default:
+5%):
+
+```
+cabal run benches -- -p /All.HashMap.lookup-miss.ByteString/ --stdev 1
+```
+
+To include comparison benchmarks for `containers` and `hashmap` uncomment the
+`cpp-options` in the benchmark section of `unordered-containers.cabal`:
+
+```
+cpp-options: -DBENCH_containers_Map -DBENCH_containers_IntMap -DBENCH_hashmap_Map
+```
+
+### References
+
+* [Documentation for `cabal`](https://cabal.readthedocs.io/en/latest/)
+* [Documentation for our testing framework, `tasty`](https://github.com/UnkindPartition/tasty#readme)
+* [Documentation for our benchmark framework, `tasty-bench`](https://github.com/Bodigrim/tasty-bench#readme)

README.md

@@ -5,6 +5,8 @@ performance critical use, both in terms of large data quantities and high speed.
 
 The declared cost of each operation is either worst-case or amortized, but
 remains valid even if structures are shared.
-  
-For developer and contributor documentation see the
+
+For background information and design considerations on this package see the
 [Developer Guide](docs/developer-guide.md).
+
+For practical advice for contributors see [`CONTRIBUTING.md`](CONTRIBUTING.md).