Avoid discrepancies between Double generation and GHC versions

Something has changed from ghc-9.0 to ghc-9.2 that often results in
generation of slightly different `Double` values. This commit carefully
chooses initial seed in such a way that generated values are exactly
the same throughout all supported ghc versions.

src/System/Random.hs

@@ -340,11 +340,11 @@ class Random a where
   -- closed interval /[lo,hi]/, together with a new generator. It is unspecified
   -- what happens if /lo>hi/, but usually the values will simply get swapped.
   --
-  -- >>> let gen = mkStdGen 2021
+  -- >>> let gen = mkStdGen 26
   -- >>> fst $ randomR ('a', 'z') gen
-  -- 't'
-  -- >>> fst $ randomR ('z', 'a') gen
-  -- 't'
+  -- 'z'
+  -- >>> fst $ randomR ('a', 'z') gen
+  -- 'z'
   --
   -- For continuous types there is no requirement that the values /lo/ and /hi/ are ever
   -- produced, but they may be, depending on the implementation and the interval.
@@ -353,8 +353,8 @@ class Random a where
   -- defined on per type basis. For example product types will treat their values
   -- independently:
   --
-  -- >>> fst $ randomR (('a', 5.0), ('z', 10.0)) $ mkStdGen 2021
-  -- ('t',6.240232662366564)
+  -- >>> fst $ randomR (('a', 5.0), ('z', 10.0)) $ mkStdGen 26
+  -- ('z',7.27305019146949)
   --
   -- In case when a lawful range is desired `uniformR` should be used
   -- instead.

src/System/Random/Stateful.hs

@@ -299,16 +299,16 @@ withMutableGen_ fg action = thawGen fg >>= action
 -- ====__Examples__
 --
 -- >>> import System.Random.Stateful
--- >>> let pureGen = mkStdGen 137
+-- >>> let pureGen = mkStdGen 139
 -- >>> g <- newIOGenM pureGen
 -- >>> randomM g :: IO Double
--- 0.5728354935654512
+-- 0.33775117339631733
 --
 -- You can use type applications to disambiguate the type of the generated numbers:
 --
 -- >>> :seti -XTypeApplications
 -- >>> randomM @Double g
--- 0.6268211351114488
+-- 0.9156875994165681
 --
 -- @since 1.2.0
 randomM :: forall a g m. (Random a, RandomGen g, FrozenGen g m) => MutableGen g m -> m a